Volume 12 - Issue 4

AI-DRIVEN TOURIST SAFETY AND RETURN COMPLIANCE SYSTEM USING GEO-FENCING AND BLOCKCHAIN DIGITAL IDENTITY

The Burgeoning Tourism Industry Has Led To Significant Vulnerabilities In The Safety Of Tourists Once They Arrive In New Destinations. Current Systems Are Based On Manual Checkins, Basic GPS Trackers Or Reactive Helpdesks - None Of Which Offer The Re... The Burgeoning Tourism Industry Has Led To Significant Vulnerabilities In The Safety Of Tourists Once They Arrive In New Destinations. Current Systems Are Based On Manual Checkins, Basic GPS Trackers Or Reactive Helpdesks - None Of Which Offer The Real-time Intelligence Or Robust Identity Protection Required In Today’s Digital World. This Paper Describes An AI-Driven Tourist Safety And Return Compliance System That Integrates Three Complementary Technologies: Blockchain-based Digital Identity, Automated Geo-fencing, And A Bespoke Transformer-Based Temporal Risk Analyzer (T-TRA). Tourists Are Allocated With Cryptographically Secure Digital Identities That Are Rooted On An Uncompromising Blockchain Network. An Intelligent Geo-fencing System Tracks Their Movements Within Safe, Prohibited And Return Zones, With Real-time Non-compliance Alerts. The T-TRA Model Ingests Temporal GPS Data Via A Self-attention Mechanism To Define The Sequence Of Safety Levels Safe, Delayed And HighRisk. In The Event Of Danger, Automatic Warnings Are Sent To Authorities And Contacts. Testing Showed All Ten Key Test Cases Were Successful. The T-TRA Model Is Scalable For Smart Tourism At Regional And National Levels And Is A Major Advance In Tourist Safety Management From Reactive To Proactive. Read more Read less

MediTrust: An AI-Based Medical Fund Verification System For Fraud Detection And Donor Trust Enhancement

This Document Presents A Demonstration Paper For Submissions To The SET International Journal Of Broadcast Engineering (SET IJBE). It Provides Authors With A Practi- Cal Example Of An AI-based Medical Fund Verification System, Including The Integrati... This Document Presents A Demonstration Paper For Submissions To The SET International Journal Of Broadcast Engineering (SET IJBE). It Provides Authors With A Practi- Cal Example Of An AI-based Medical Fund Verification System, Including The Integration Of Deep Learning Models, Document Processing Techniques, And Fraud Detection Mechanisms. The Paper Illustrates The Structure Of A Complete Manuscript, Covering Title, Author Information, Abstract, Keywords, Sections, Figures, Tables, Equations, Lists, References, And Acknowledgments. The Proposed System, MediTrust, Utilizes CRAFT For Text Region Detection, Donut For Document Understanding And Structured Data Extraction, And Fuzzy Matching Algorithms For Validating Ex- Tracted Information Against A Trusted Hospital Database. The Sam- Ple Content Demonstrates How Automated Verification Improves Transparency, Reduces Manual Effort, And Detects Fraudulent Medical Fund Requests Efficiently. This Abstract Is Limited To Fewer Than 150 Words, In Accordance With The Journal Guidelines, And Is Intended To Guide Authors In Preparing Compliant, Consistent, And Publication-ready Submissions. It May Also Serve As A Reference When Adapting The Template To Real-world AI-based Research Contributions. Read more Read less

ONLINE VOTING SYSTEM USING FACE RECOGNITION AND OTP (ONE- TIME PASSWORD)

The Main Purpose Of The System Is To Create An Online Voting Portal. This Portal Addresses The Drawbacks Of Manual Voting And Earlier Online Voting Systems Using Facial Recognition And OTP Generation. People Can Vote From Anywhere If They Are Unable... The Main Purpose Of The System Is To Create An Online Voting Portal. This Portal Addresses The Drawbacks Of Manual Voting And Earlier Online Voting Systems Using Facial Recognition And OTP Generation. People Can Vote From Anywhere If They Are Unable To Attend Their Designated Voting Booths, Providing Location Flexibility. The System Ensures Security For Voters And Administrators With Several Authentication Layers. These Include Facial Recognition And OTP Authentication Using Registered User Details. A Person Can Use The Online System Only After Their ID Is Verified Against The Database Of Registered Voters. Read more Read less

Farmnet : Smart Farming

Agriculture Plays A Major Role In The Indian Economy And Is The Primary Source Of Livelihood For Many People. However, Farmers Often Face Several Challenges Such As Lack Of Proper Crop Guidance, Fluctuating Market Prices, Sudden Weather Changes, Inco... Agriculture Plays A Major Role In The Indian Economy And Is The Primary Source Of Livelihood For Many People. However, Farmers Often Face Several Challenges Such As Lack Of Proper Crop Guidance, Fluctuating Market Prices, Sudden Weather Changes, Incorrect Use Of Fertilizers, And Limited Access To Modern Digital Tools. These Problems Can Affect Productivity And Income. To Address These Issues, A Smart Web-based Platform Named FARMNET Has Been Developed. FARMNET Is Designed To Support Farmers By Providing Multiple Services In A Single System. It Offers Crop Recommendations Based On Farming Needs, Regular Weather Updates, Fertilizer Suggestions, Basic Disease Identification Guidance, Current Market Rates, And Expert Assistance. By Combining These Features, The Platform Helps Farmers Make Better Farming Decisions. The System Is Developed Using Technologies Such As Node.js, Express.js, MongoDB Atlas, HTML, CSS, And JavaScript. It Can Also Be Integrated With A Spring Boot Java Backend For Advanced Functionalities And Future Expansion. Farmers Can Create Accounts, Log In Securely, Save Their Data. Read more Read less

Multi-Feature Search–Based Purchasing Tendency Community Classification For Densely Distributed Clients In E-commerce

Purchasing Tendency Is Defined As Customer Prefer-ences For Products And Brands, Interested In Price And Frequency Of Purchase, And Is Determined By Demographic, Transactional And Behavioral Attributes. In Today’s E-commerce, These Insights Are Criti... Purchasing Tendency Is Defined As Customer Prefer-ences For Products And Brands, Interested In Price And Frequency Of Purchase, And Is Determined By Demographic, Transactional And Behavioral Attributes. In Today’s E-commerce, These Insights Are Critical For Recommendations And Managing Demand. But Conventional If-then Rules And Elementary Collaborative Filtering Approaches Lack Sophisticated Insights Into Interactions Between Customers, Products, And Locations, And The Demand At Different Times. To Overcome These Challenges, This Article Proposes A Community Classification System Of Clients Purchasing Inventory Using A Holistic Deep Learning Approach. Graph Neural Networks (GNNs) Capture The Relationship Between Customers, Products, And Regions, Facilitating Precise Identification Of Customer Com-munities And Region-based Demand (high, Emerging, Low). The SASRec Transformer-based Model Also Leverages Temporal In-formation About Customer Preferences By Training On Temporal Sequences, Capturing Both Short- And Long-term Information. This Approach Incorporates Demographic, Transactional And Be-havioral Data To Offer Insights To Sellers And Recommendations To Customers, Thus Improving Decision-making, Forecasting, And Efficiency In The Market. Read more Read less

ADAPTIVE GAN ASSISTED ENCRYPTION MODEL FOR SECURITY ENHANCEMENT IN BLOCKCHAIN BACKED TOKENIZED DIGITAL ASSETS

In The Present-day Age, Where We Are Highly Dependent On Technology, The Safe Transmission Of Confidential Or Personal Images Through The Internet Has Become Very Important. The Conventional Way Of Transmitting Or Sharing Images Is Done Through The U... In The Present-day Age, Where We Are Highly Dependent On Technology, The Safe Transmission Of Confidential Or Personal Images Through The Internet Has Become Very Important. The Conventional Way Of Transmitting Or Sharing Images Is Done Through The Use Of Basic Authentication Protocols And Individual Encryption Methods, Which Might Not Be Enough To Combat Cyber-attacks And Other Dangers. In Order To Overcome These Problems, The Proposed System, Named SECUREGAN: Hybrid ECC-AES Encrypted NFT-Based Image Sharing System, Will Implement A Multilayer Security System, Combining Blockchain Technology For NFT Verification, GANs, And Encryption. In This Process, The User Registers Himself And Logs In To Transmit The Image As An NFT. Furthermore, GAN Is Employed To Create A Random Image That Serves As A Camouflage Layer, Thus Making It Extremely Difficult For An Attacker To Be Able To Decipher The Information Contained In The Image. Additionally, In Order To Increase The Security Of The Scheme, It Uses The Original Image And Combines It With The Random Image Generated Using The GAN Model. Then, The Data Is Split Into Multiple Shares. Each Of These Shares Is Subjected To Hybrid Encryption Using AES And ECC Cryptography. AES Is Used For Data Encryption, While ECC Is Used For Key Exchange. In This Manner, The Encrypted Shares Are Delivered To The Recipient Over Untrusted Channels. On Reception, The Receiver Uses The Cryptographic Key In Order To Decrypt The Shares, Merge Them, And Finally Reconstruct Both The Original And Random Images. Read more Read less

AN INTELLIGENT MULTI-ALGORITHM MACHINE LEARNING SYSTEM FOR SKIN DISEASE CLASSIFICATION AND PREDICTION

This Paper Proposes An Interpretable And Explainable Multi-class Skin Lesion Classification Model For The HAM10000 Dataset. Timely Detection Of Skin Cancer Is Important But Difficult Due To Class Imbalance And Uninterpretability Of Current Models. Th... This Paper Proposes An Interpretable And Explainable Multi-class Skin Lesion Classification Model For The HAM10000 Dataset. Timely Detection Of Skin Cancer Is Important But Difficult Due To Class Imbalance And Uninterpretability Of Current Models. The Model Employs EfficientNetV2-L With Channel Attention For Improved Feature Extraction And Classification. An Effective Preprocessing Strategy With Data Augmentation And Smart Class Balancing Is Used. Progressive Training In Three Stages Enhances Generalization. Visual Explainable AI, Including Grad-CAM And Saliency Maps, Explains Predictions Visually. The Model Has 91.15% Accuracy And 99.33% AUC Across Seven Classes. The Solution Enhances Diagnostic Performance And Interpretability, Enabling Its Use As A Decision Support System. Read more Read less

SheG: The Gridless Guardian – An Autonomous Offline-First Safety Ecosystem For Women

Women’s Safety Continues To Remain A Major Social Concern Across The World. Existing Safety Applications Mostly Depend On Continuous Internet Connectivity, Which Becomes A Serious Limitation During Emergencies Occurring In Rural Regions, Underground... Women’s Safety Continues To Remain A Major Social Concern Across The World. Existing Safety Applications Mostly Depend On Continuous Internet Connectivity, Which Becomes A Serious Limitation During Emergencies Occurring In Rural Regions, Underground Areas, Or Crowded Places With Unstable Networks. To Address This Issue, This Paper Introduces SheG: The Gridless Guardian, An Autonomous Offline-first Women’s Safety Ecosystem Designed To Provide Reliable Emergency Support Even Without Internet Access. The Proposed System Uses Built-in Smartphone Technologies Such As GPS, GSM-based SMS Communication, Biometric Authentication, Microphone, And Camera Modules To Deliver Emergency Alerts, Evidence Recording, And Secure Access Control. Unlike Conventional Safety Applications, SheG Ensures Uninterrupted Emergency Communication Through GSM SMS Services That Operate Independently Of Mobile Data Connectivity. Experimental Evaluation Conducted Under Different Network Conditions Demonstrated A 97.8% SMS Delivery Success Rate In Low-network Environments, Significantly Outperforming Conventional Internet-dependent Systems. The Application Achieved An Average SOS Response Time Of 3.7 Seconds With Minimal Battery Consumption During Continuous Background Monitoring. The Proposed Platform Provides A Practical, Scalable, And Highly Reliable Solution For Women’s Safety, Especially In Resource Constrained And Low-connectivity Environments. Read more Read less

Wearable Non-invasivs Glucose Monitoring Using Optical Sensing

Non-invasive Glucose Monitoring Remains A Critical Challenge In Biomedical Engineering Due To The Complex Interaction Of Light With Biological Tissues And The Weak Specificity Of Glucose-related Optical Signals. This Paper Presents The Design And Imp... Non-invasive Glucose Monitoring Remains A Critical Challenge In Biomedical Engineering Due To The Complex Interaction Of Light With Biological Tissues And The Weak Specificity Of Glucose-related Optical Signals. This Paper Presents The Design And Implementation Of A Wearable Non-invasive Glucose Monitoring System Based On Near-Infrared (NIR) Optical Sensing. The Proposed System Utilizes A 940 Nm NIR Light Source, Selected Within The Optical Window That Minimizes Water Absorption While Ensuring Adequate Penetration Into The Skin. The Sensing Principle Is Based On The Beer–Lambert Law, Where The Attenuation Of Light Intensity Is Related To The Absorption Coefficient Of The Medium, Which Is Influenced By Glucose Concentration. A Photodiode Detects The Diffusely Reflected Light, And The Resulting Signal Is Amplified, Filtered, And Digitized Using An ESP32 Microcontroller. Signal Preprocessing Techniques Are Applied To Reduce Noise And Improve Measurement Stability. The Processed Data Is Transmitted Wirelessly To A Mobile Application Via Bluetooth Low Energy (BLE) For Real-time Monitoring. Due To The Presence Of Significant Optical Scattering, Physiological Variability, And Dominant Water Absorption, The Extracted Signal Represents A Composite Response Rather Than A Glucose-specific Measurement. Therefore, The Current System Focuses On Demonstrating Feasibility Rather Than Achieving Clinical-grade Accuracy. Experimental Results Confirm Successful Signal Acquisition And Wireless Transmission, Validating The System Architecture. Future Work Will Involve Multi-wavelength Sensing, Advanced Signal Processing, And Machine Learning-based Calibration Models To Improve Accuracy And Robustness. This Study Contributes Toward The Development Of Practical, Wearable, And Non-invasive Glucose Monitoring Technologies. Read more Read less

AI-Based Online Interview Behavioral Analysis: An AI-Driven Approach To Proctoring And Integrity Verification

The Rapid Proliferation Of Remote Work Culture And Digitally-driven Recruitment Ecosystems Has Fundamentally Reshaped How Organizations Identify And Evaluate Talent. Online Interviews Have Emerged As The Dominant Mode Of Candidate Assessment, Offerin... The Rapid Proliferation Of Remote Work Culture And Digitally-driven Recruitment Ecosystems Has Fundamentally Reshaped How Organizations Identify And Evaluate Talent. Online Interviews Have Emerged As The Dominant Mode Of Candidate Assessment, Offering Logistical Convenience And Geographic Flexibility. However, This Shift Introduces Unprecedented Vulnerabilities In Maintaining The Integrity And Fairness Of The Hiring Process. Candidates Operating In Unmonitored Remote Environments May Exploit The Absence Of Physical Supervision By Consulting Hidden Reference Materials, Receiving Real-time Verbal Or Textual Coaching From Off-camera Individuals, Utilizing Unauthorized Secondary Displays, Or Leveraging AI-based Answer Generation Tools — All Of Which Fundamentally Compromise The Validity Of The Assessment. This Paper Presents The Online Interview Behavioral Analysis (OIBA) System, A Comprehensive, Privacy-first, AI-driven Behavioral Monitoring Framework Specifically Engineered To Detect And Quantify Integrity Violations During Virtual Job Interviews. Unlike Existing Academic Proctoring Tools That Rely On Cloud-based Video Surveillance Or Simplistic Rule-based Anomaly Detection, OIBA Employs A Multi-modal Micro-module Architecture That Operates Entirely On-device, Ensuring Candidate Data Confidentiality While Delivering Real-time Analytical Precision. The System Integrates Four Parallel Analytical Pipelines. The Visual Analysis Subsystem Leverages Google MediaPipe Face Mesh To Extract 468 Three-dimensional Facial Landmarks Per Frame, Enabling Precise Computation Of Iris Displacement Vectors For Gaze Direction Classification Across Five Categories: Center, Left, Right, Up, And Down. OpenCV's Perspective-n-Point (solvePnP) Algorithm Processes These Landmarks To Compute Pitch, Yaw, And Roll Rotation Vectors, Enabling Head Pose Estimation With Sub-degree Angular Resolution. The Affective Analysis Component Employs DeepFace For Real-time Facial Emotion Recognition, Classifying Expressions Into Seven Categories And Correlating Elevated Stress Or Fear Indicators With Concurrent Behavioral Anomalies To Strengthen Detection Confidence. The Audio Intelligence Subsystem Utilizes Picovoice Falcon For On-device Speaker Diarization, Separating Audio Streams Into Distinct Speaker Tags Without Transmitting Sensitive Data To External Servers, While Picovoice Leopard Performs Speech-to-text Transcription For Subsequent NLP-based Content Matching Between Secondary Speaker Prompts And Candidate Responses. All Four Modules Contribute Weighted Signals To A Unified Suspicion Score (0–100) Rendered On A Flask-based Web Dashboard Via JustGage Visualization. A Key Innovation Of The System Is Its Adaptive Threshold Mechanism, Which Requires Continuous Anomalous Behavior For A Minimum Duration (2 Seconds For Visual Violations, 500ms For Audio Anomalies) Before Registering A Suspicion Event — A Design Decision That Reduced False Positive Rates By 62% In Controlled Testing Without Degrading Sensitivity To Genuine Violations. Experimental Evaluation On 150 Simulated Interview Sessions Demonstrated An Overall Detection Accuracy Of 89%, With Individual Module Accuracies Of 94.2% (eye Gaze), 91.3% (head Pose), 83.1% (emotion Recognition), And 92% (audio Diarization). The System Processes A 30-minute Interview In Approximately 138 Seconds, Representing A 4.4x Speed-up Relative To Real-time. Comparative Analysis Confirms That OIBA Outperforms Existing Commercial Proctoring Solutions Across Accuracy, Latency, And Privacy Preservation Metrics. The OIBA System Demonstrates That Multi-modal Behavioral AI, When Designed With A Privacy-first, On-device Processing Philosophy, Can Serve As A Reliable, Scalable, And Non-intrusive Solution For Maintaining The Integrity Of Remote Hiring Processes — Providing Recruiters With Evidence-backed, Actionable Insights Rather Than Subjective Human Judgment Alone. Read more Read less

HOLISTIC SMARTPHONE DATA PROTECTION SYSTEM INTEGRATING ANDROID APP ANALYSIS AND SECURE METADATA TRACKING

This Project Aims To Design A *Comprehensive Smartphone Data Security System* To Enhance Security Of Android Phone By Smart Malware Detection And Metadata Management. With The Increased Use Of Mobile Devices For Critical Applications Such As Telephon... This Project Aims To Design A *Comprehensive Smartphone Data Security System* To Enhance Security Of Android Phone By Smart Malware Detection And Metadata Management. With The Increased Use Of Mobile Devices For Critical Applications Such As Telephony And Banking, Traditional Signaturebased Security Tools Are No Longer Effective To Identify Advanced Types Of Malware Like Zero-day And Polymorphic Attacks. Our System Addresses These Issues By Employing A Machine Learning Based Model Using XGBoost Algorithm To Detect Malware Android Apps By Considering Both Static And Dynamic Features, Such As Permissions, API Calls And Usage Patterns. This Enables Efficient Detection Of Malicious Apps With Minimal False Positives. In Addition, We Implement A Blockchain-like JSON-based Module To Ensure Data Integrity And Transparency, By Storing The Detection Results And Information About The Model. This Guarantees Integrity And Allows Pre-installation Authentication Of Apps. Our System Also Incorporates Other Important Components Such As A Malware Level Indicator And A Proactive Blocker To Prevent The Installation Or Distribution Of Critical Applications. Overall, Our System Provides A Scalable, Real-time And Efficient System To Improve Mobile Security Against The Ever-evolving Threats. Read more Read less

Crowdsourced Civic Issue Reporting And Resolution: A Digital Platform For Transparent Urban Governance

Urban Municipal Infrastructure In India Faces A Lot Of Issues Such As Potholes, Improper Water Management, Broken Streetlights, Water Leakage And Drainage Blockages. The Major Challenge That We Are Facing In This Modern Era Is That The Authorities De... Urban Municipal Infrastructure In India Faces A Lot Of Issues Such As Potholes, Improper Water Management, Broken Streetlights, Water Leakage And Drainage Blockages. The Major Challenge That We Are Facing In This Modern Era Is That The Authorities Delay These Kinds Of Problems Nor Completely Ignore It. The Existing Complaint Methods Are Visiting The Offices Or Calling The Helplines, Which Is Time Consuming And Inconvenient. Our Project Proposes A Crowdsourced Civic Issue Reporting And Resolution System, This Is A Digital Platform Which Was Developed Using The Python Flask For The Backend, React.js For The Web Interface, And React Native For The Mobile Application. In This Platform, It Allows The Users To Report The Issue By Capturing The Problem Or Issue And Attaching That Image Into This Platform With The Location Details Which Will Be Detected By The App Automatically, Selecting The Category And Submitting The Complaint In A Simple Manner. The Problem Will Be Identified By The AI. Here, The System Uses Machine Learning To Read The Description That We Have Given And It Analyses The Image Which Was Attached And Decides Which Department Should Fix The Particular Problem. It Allows Automatic Identification Of The Issue And Routes To The Appropriate Municipal Department. The Complaint Goes Through Four Stages: Reported → In Progress → Considered → Resolved. The Notification Will Be Sent Directly To The User After The Problem Is Resolved. The Performance Of This Platform Is That It Works Faster. Many People Can Use It Simultaneously, And Notification Reaches Almost Everyone Successfully. Testing Results Indicate That 92% Of People Can Submit The Complaint Successfully On Their First Try. Also, The Official Has Tested And Worked Well. This Scalable Solution Can Significantly Improve The Civic Issue Management And Enhance The Communication Between The Citizens And Authorities. Read more Read less

AI-Powered Inventory Management System- Ezze Buy

This Paper Presents The Design, Architecture, And Full-stack Implementation Of EzzeBuy, A Web-based AI-powered Inventory Management And Sales Prediction Platform. The System Integrates A Long Short-Term Memory (LSTM) Neural Network Backend For Dynami... This Paper Presents The Design, Architecture, And Full-stack Implementation Of EzzeBuy, A Web-based AI-powered Inventory Management And Sales Prediction Platform. The System Integrates A Long Short-Term Memory (LSTM) Neural Network Backend For Dynamic Sales Forecasting, A Flask-based REST API For Inventory Operations, CSV-driven Data Ingestion With Drag-and-drop Support, A Data Layer Managed Using Pandas And CSV Persistence, And A Responsive Frontend Built With HTML5, CSS3, And JavaScript. The Platform Supports Real- Time Dashboard KPI Tracking, Low-stock And Near-expiry Alerting, Product-level Analytics, And AI-powered Demand Forecasting With Configurable Prediction Horizons. The Proposed Architecture Provides A Reproducible Foundation For Developing Scalable AI- Enabled Inventory Management Platforms Suitable For Small And Medium Enterprises. Read more Read less

Autonomous Agri Bots For Field Surveillance

Modern Agriculture Increasingly Requires Intelligent Systems Capable Of Continuous Monitoring And Data-driven Decision Support To Improve Crop Yield And Sustainability. This Report Presents A Raspberry Pi-based Intelligent Autonomous Farming Robot De... Modern Agriculture Increasingly Requires Intelligent Systems Capable Of Continuous Monitoring And Data-driven Decision Support To Improve Crop Yield And Sustainability. This Report Presents A Raspberry Pi-based Intelligent Autonomous Farming Robot Designed For Real-time Plant Health Monitoring, Disease Analysis, And Fertilizer Application. The System Integrates A Suite Of Hardware Including A Soil Moisture Sensor And A Raspberry Pi Camera Module To Collect Comprehensive Environmental And Visual Data. Using Image Processing Algorithms, The Raspberry Pi Analyzes Leaf Characteristics Such As Color, Texture, And Spots - To Identify Common Crop Diseases And Stress Levels At An Early Stage. It Includes The Robotic Platform Systematically Traverse Agricultural Fields With Minimal Human Intervention, Sensor Data, Health Status, And Specific Fertilizer Application Are Transmitted To A Remote Platform Or Mobile App For Farmer Action, By Optimizing Irrigation And Reducing Fertilizer Misuse, The System Provides A Cost-effective And Scalable Solution For Smart Farming. Read more Read less

AI-Based Quiz Solving Tutor With Personalized Learning

This Work Introduces The Design, Architecture, And Full-stack Development Of Quizmify Production, An AI-based SaaS Grade Quiz Creation And Real-time Multiplayer Assessment Appli-cation. The Solution Includes An LLM-based Question Generator Backend (G... This Work Introduces The Design, Architecture, And Full-stack Development Of Quizmify Production, An AI-based SaaS Grade Quiz Creation And Real-time Multiplayer Assessment Appli-cation. The Solution Includes An LLM-based Question Generator Backend (Groq – Llama-3.3-70B), Socket.IO-driven Real-time Multiplayer Environment For Engaging Quizzes, OAuth2 Authenti-cation Via NextAuth.js, PostgreSQL Database With Prisma ORM Management, And A Razorpay Payment Gateway For Managing Premium Subscriptions. The Web Application Is Built Using Next.js 15, React 19, And TypeScript 5.8, With Support For Both Multiple-choice And Free Response Quizzes With Adjustable Difficulty, Topic Filtering Options, And An NLP-based Keyword Overlap Evaluation Approach For Short Answers. Experiments Reveal Sub-200 Ms Question Generation Time, Under 20 Ms Answer Evaluation Time, And Stable Real-time Multiplayer Communication With Around 500 Concurrent Rooms Per Node.js Instance. The Architectural Design Offers A Reproducible Framework For Building Scalable And AI-integrated Educational Applications. Read more Read less

Pujanam: A Comprehensive Portal For Pandit Booking, Puja Services & Samagri Management

This Paper Introduces Pujanam, A Web-based Plat-form That Consolidates Pandit Booking, Puja Scheduling, And Samagri Procurement Into A Single Digital Workflow. Arrang-ing Religious Services Through Conventional Means—phone Calls, Word-of-mouth Referr... This Paper Introduces Pujanam, A Web-based Plat-form That Consolidates Pandit Booking, Puja Scheduling, And Samagri Procurement Into A Single Digital Workflow. Arrang-ing Religious Services Through Conventional Means—phone Calls, Word-of-mouth Referrals, And In-person Visits—often Leads To Scheduling Conflicts, Price Uncertainty, And Limited Reach For Both Devotees And Pandits. Pujanam Targets This Operational Gap Through A Role-differentiated System Serving Devotees, Pandits, And Administrators. Devotees Can Browse A Categorized Puja Catalog, Inspect Pandit Profiles Complete With Ratings And Experience Summaries, Select Available Time Slots, And Place Samagri Orders. Pandits Access A Scheduling Dashboard, Receive Live Booking Alerts Via Socket.IO, And Maintain A Transaction History. Administrators Oversee The Full Platform Through A Dedicated Control Panel Covering User Management, Service Listings, And Support Resolution. The Backend Is Built On Node.js/Express With JWT Authen-tication, Bcrypt Password Hashing, And MongoDB For Schema-flexible Storage. Socket.IO Drives The Event-based Notification Layer. Together, These Components Deliver A Cohesive System That Lowers Coordination Friction And Broadens Access To Religious Services For Urban Communities. Read more Read less

DESIGN AND ANALYSIS OF RAMJET ENGINE FOR SUPERSONIC FLIGHT

The Increasing Demand For High-speed Aircraft And Missile Systems Has Led To The Development Of Propulsion Systems Capable Of Operating Efficiently At Supersonic Speeds. This Study Focuses On The Design And Computational Analysis Of A Ramjet Engine I... The Increasing Demand For High-speed Aircraft And Missile Systems Has Led To The Development Of Propulsion Systems Capable Of Operating Efficiently At Supersonic Speeds. This Study Focuses On The Design And Computational Analysis Of A Ramjet Engine Inlet For Supersonic Flight Applications, Specifically Examining The Influence Of Cone Angle On Pressure Recovery Performance. Three Inlet Cone Angles (43°, 44°, And 45°) Were Modeled Using SolidWorks And Subjected To Cold-flow Analysis In ANSYS Fluent At Mach Numbers 2, 2.5, And 3. The K-epsilon Turbulence Model Was Employed With Density-based Solver Settings Under Steady-state Conditions. Results Indicate That A 45° Cone Angle Achieves Maximum Pressure Recovery At Mach 3, Producing An Exit Pressure Of 1,585,917.50 Pa And Generating A Shock-train Configuration Favorable For Pre-combustion Conditions. The 44° Cone Angle Demonstrates The Highest Absolute Exit Pressure (2,806,599.25 Pa) At Mach 3 But Requires Supercritical Inlet Conditions. The Findings Provide Design Guidance For Artillery Ramjet Inlet Geometry Optimization And Contribute To The Broader Understanding Of Supersonic Air-breathing Propulsion Systems. Read more Read less

Effectiveness Of A Gamified Learning Approach Compared To Google Form–Based Assessment In Medical Education: Randomised Control Study

Background: Gamification Is Increasingly Used In Medical Education To Improve Learner Engagement, Motivation, And Knowledge Retention.There Is A Worldwide Shortage Of Health Workers, And This Issue Requires Innovative Education Solutions. Serious Gam... Background: Gamification Is Increasingly Used In Medical Education To Improve Learner Engagement, Motivation, And Knowledge Retention.There Is A Worldwide Shortage Of Health Workers, And This Issue Requires Innovative Education Solutions. Serious Gaming And Gamification Education Have The Potential To Provide A Quality, Cost-effective, Novel Approach That Is Flexible, Portable, And Enjoyable And Allow Interaction With Tutors And Peers.[1]However, Evidence Comparing Gamified Strategies With Conventional Google Form–based Assessments Is Limited. Objectives: To Evaluate The Effectiveness Of A Gamified Learning Approach Compared To Standard Google Form Quizzes Among Undergraduate Medical Students. Methods: A Randomised Control Trial Was Conducted Among Second-year MBBS Students Using Computerised Randomisation. Participants Were Divided Into Two Groups: (1) Google Form Group, Using Traditional Multiple-choice Assessments And (2) Gamified Learning Group, Using An Interactive Platform (e.g., Kahoot/Quizizz). Pre- And Post-tests Were Administered. Outcomes Included Improvement In Test Scores, Student Engagement, Satisfaction, And Perceived Usefulness Assessed Via Likert-scale Questionnaires. Data Were Analysed Using Independent Sample T-test. Results: The Parameters Measured Have Demonstrated Effectiveness; Encouraged Review (3.40), Fit With Lectures (3.65), Increased Understanding (3.59), Lower Recall (3.19), Moderate User Engagement (3.15), Interactivity (3.18) And Interest (3.14). Usability Has Been Assessed As Satisfactory: User Friendly (3.60), Good Difficulty (3.48) And Appropriate Time (3.56). Group 2 Significantly Outperform Group 1 In Every Aspect, Especially In The Recall (4.13 And 2.25), Engagement (4.13 And 2.18), Interactivity (4.18 And 2.18) And Interest (3.83 And 2.45) Parameters. The Usability Parameters In Group 2 Also Significantly Outperform Group 1: User Friendly (3.95 And 3.25), Difficulty (3.68 And 3.28) And Time (3.75 And 3.38). The Range For Standard Deviation (from 0.59 To 1.20) Represents Moderate Variance And Overall Results Indicate That Group 2 Appears More Successful. Conclusion: Gamified Learning Is More Effective Than Google Form–based Assessments In Improving Learning Outcomes, Engagement, And Satisfaction In Medical Education. Incorporation Of Gamified Tools Is Recommended For Routine Teaching-learning Activities. Read more Read less

A MACHINE LEARNING BASED CLASSIFICATION AND PREDICTION TECHNIQUES FOR DDOS ATTACKS

Distributed Denial-of-Service (DDoS) Attacks Represent A Significant Threat In Contemporary Network Security, Compromising The Availability And Integrity Of Services Across Diverse Platforms. These Attacks Overwhelm Target Networks With Substantial T... Distributed Denial-of-Service (DDoS) Attacks Represent A Significant Threat In Contemporary Network Security, Compromising The Availability And Integrity Of Services Across Diverse Platforms. These Attacks Overwhelm Target Networks With Substantial Traffic Volumes, Frequently Causing System Failures Or Rendering Services Unresponsive. As DDoS Attacks Continue To Increase In Scale And Sophistication, Conventional Detection Methodologies, Including Signature-based Systems And Threshold-based Approaches, Demonstrate Insufficient Effectiveness. These Traditional Methods Often Exhibit Elevated False Positive Rates And Detection Delays, Potentially Resulting In Considerable Damage Or Service Disruption Before Remedial Actions Can Be Taken. To Overcome These Limitations, This Paper Presents The Development Of An Adaptive Detection System (ADS) For Identifying And Mitigating Network DoS And DDoS Attacks. The Proposed System Employs Advanced Sampling Techniques And Machine Learning (ML) Algorithms To Perform Dynamic Network Traffic Analysis And Achieve More Precise Identification Of Malicious Patterns. In Contrast To Conventional Approaches, The Proposed System Demonstrates The Capability To Adapt To The Continuously Evolving Landscape Of Cyberattacks, Thereby Minimizing The Probability Of Undetected Attacks Or False Positives. The Research Concentrates On Determining The Upper Bounds Of DoS Attack Frequency And Duration, Particularly The Threshold Parameters At Which Systems Can Withstand Attacks While Maintaining Network Consensus. Through The Integration Of Adaptive Detection Capabilities, Reduced Computational Complexity, And Enhanced System Resilience, This Approach Presents A Viable Solution For Protecting Networks Against Increasingly Sophisticated DDoS Attacks. Read more Read less

DropLine: An Asynchronous RAG Architecture For Resolving Wrapper Link Bottlenecks

We Present DropLine, An AI-assisted System That Converts Arbitrary Web Links Into Structured Educational Content Using Retrieval-augmented Generation (RAG). The System Addresses The Problem Of “signpost Versus Destination” Links, Where Many URLs Func... We Present DropLine, An AI-assisted System That Converts Arbitrary Web Links Into Structured Educational Content Using Retrieval-augmented Generation (RAG). The System Addresses The Problem Of “signpost Versus Destination” Links, Where Many URLs Function Only As Pointers To Boilerplate HTML While Their Intended Information Remains Hidden Behind Dynamic Frames, Redirects, Or Embedded Media. We Formalize This As A URI Resolution Challenge Combined With Dynamic Content Accessibility. DropLine’s Backend Is Implemented With FastAPI For Asynchronous Throughput. It Resolves URLs, Follows Redirects And Shorteners, Detects Content Type Such As Webpage, Video, Or Image, And Extracts Content Using Specialized Modules. For Web Pages, We Use Trafilatura To Perform High-precision Boilerplate Removal; For Videos, We Use The Youtube-transcript-api To Retrieve Captions Without Requiring An API Key. The Cleaned Text, Such As A Wikipedia Article Or Video Transcript, Is Then Passed To Google DeepMind’s Gemini 2.5 Flash Model, Which Supports A Large Multimodal Context Window. Our Prompt Instructs Gemini To Generate Five Pedagogical Outputs: A Concise Summary, Key Concepts, An Analogy-based Explanation For Beginners, Real-world Applications, And A Short Quiz. A Streamlit User Interface Presents The Result As An Interactive Tutor. Using St.session_state, We Cache The Document Context And Conversation History, Allowing Users To Ask Follow-up Questions Grounded In The Same Source Material. In Evaluation, DropLine Successfully Processed Over 1.2 Million Characters Of Raw HTML From A Wikipedia Case Study And Produced Approximately 180,232 Characters Of Clean Text, Retaining About 15% Of The Original Content, Which Is Consistent With Known Benchmarks. The System Also Recovered From HTTP 503 And 429 Errors Using An Exponential Backoff Strategy. These Results Show That The Asynchronous RAG Pipeline Can Reliably Bridge Modern Web Links And Rich Multimodal Knowledge Deliver. Read more Read less

Improving Drying Efficiency Of Direct Type Solar Dryer Using Gen AI

Solar Drying Is A Sustainable, Eco-friendly, And Cost-effective Method Used For Preserving Agricultural And Food Products By Utilizing Renewable Solar Energy As An Alternative To Conventional Fuel-based Drying Methods. However, Traditional Direct Typ... Solar Drying Is A Sustainable, Eco-friendly, And Cost-effective Method Used For Preserving Agricultural And Food Products By Utilizing Renewable Solar Energy As An Alternative To Conventional Fuel-based Drying Methods. However, Traditional Direct Type Solar Dryers Are Generally Fixed In Position And Unable To Follow The Movement Of The Sun, Which Results In Reduced Heat Absorption, Uneven Moisture Removal, Longer Drying Time, And Fluctuations In Chamber Temperature That May Affect The Quality Of Dried Products. To Overcome These Limitations, This Project Proposes An Improved Direct Type Solar Dryer Integrated With Automatic Sun Tracking And Generative AI Based Performance Optimization. The System Uses LDR (Light Dependent Resistor) Sensors To Detect The Direction Of Maximum Sunlight Intensity And An Arduino Uno Controller To Process Sensor Data And Operate An SG90 Servo Motor For Rotating The Dryer Towards The Sun Throughout The Day. By Maintaining Maximum Solar Exposure, The Dryer Is Capable Of Achieving Higher Internal Temperature, Faster Moisture Evaporation, And Improved Drying Efficiency. In Addition, Generative AI Techniques Are Utilized For Conceptual Design Enhancement, Drying Parameter Prediction, And Optimization Of System Performance. Phase 1 Of The Project Mainly Focuses On Design Analysis, Component Selection, Control Logic Development, And Expected Performance Evaluation Before Prototype Fabrication. The Proposed System Is Expected To Reduce Drying Time, Improve Product Quality, Minimize Manual Intervention, And Increase Energy Utilization Efficiency When Compared With Conventional Fixed Solar Dryers. This Low-cost And Practical Solution Is Highly Suitable For Small-scale Agricultural And Food Processing Applications, Especially In Rural And Remote Areas. Read more Read less

IoT Based Real Time Crop Irrigation And Energy Efficient System

Conventional Irrigation Methods Often Result In Excessive Water Use, Energy Waste, And Inadequate Fertilizer Management, Despite The Fact That Agriculture Is Crucial To Economic Prosperity. To Solve These Problems, This Study Proposes An Internet Of... Conventional Irrigation Methods Often Result In Excessive Water Use, Energy Waste, And Inadequate Fertilizer Management, Despite The Fact That Agriculture Is Crucial To Economic Prosperity. To Solve These Problems, This Study Proposes An Internet Of Things (IoT)-based Real-time Crop Irrigation And Energy-efficient System That Integrates Automatic Water Level Monitoring, Remote Load Control, And Soil Nutrient Analysis. The Proposed Solution Uses An ESP32 Microcontroller To Monitor The Water Tank's Level And Promptly Alert The Smart Water Mobile Application When The Tank Fills Up. Relay Modules Are Used To Control Electrical Loads, Such As Lightbulbs Or Irrigation Pumps, To Reduce Power Consumption And Avoid Overflow. Additionally, An NPK Soil Sensor Interfaced Via RS485 Modbus Communication Provides Real-time Potassium, Phosphorus, And Nitrogen Measurements Read more Read less

AI-Driven Banking Security Monitoring System

Banking Environments Consistently Rank Among The Most Security-critical Operational Domains, Where Institutions Must Safeguard Financial Assets, Sensitive Customer Information, And Personnel Against Theft, Fraud, And Unauthorized Access. Employees An... Banking Environments Consistently Rank Among The Most Security-critical Operational Domains, Where Institutions Must Safeguard Financial Assets, Sensitive Customer Information, And Personnel Against Theft, Fraud, And Unauthorized Access. Employees And Customers Operate Within Dynamic Indoor Spaces Where Incidents Such As Suspicious Movement, Identity Concealment, Or Unauthorized Entry May Occur Undetected, Especially Under Low Lighting, Occlusions, And Crowded Conditions That Conventional Surveillance Systems Cannot Reliably Analyse In Real Time. Existing Solutions — Passive CCTV Monitoring, Rule-based Motion Detection, And Continuous Manual Supervision — Each Fail To Deliver Autonomous, Accurate, And Real-time Threat Detection Across Complex And High-traffic Banking Environments. This Paper Introduces AI-Bank Secure, A Vision-based Anomaly Detection And Facial Recognition Framework Purpose-built To Address These Limitations. The System Continuously Processes Surveillance Camera Video Through A Motion Vector-based Analysis Pipeline Integrated With A Gaussian Mixture Model (GMM), Extracting Foreground Segmentation, Object Motion Patterns, And Behavioural Deviations To Identify Suspicious Activities. A Tracking Module Employing Blob Analysis And Inter-frame Object Association Reliably Monitors Movement Trajectories And Distinguishes Abnormal Behaviours Such As Loitering Or Sudden Directional Changes From Normal Customer Activity. Simultaneously, An ArcFace-based Deep Learning Model Generates Discriminative Facial Embeddings To Perform Real-time Identity Verification Against A Registered Database.Upon Confirmed Detection Of Anomalous Behaviour Or Unidentified Individuals, Multi-channel Security Alerts Are Dispatched Immediately, Including Captured Evidence Frames, System Notifications, And Automated Messages To Security Personnel — All Without Human Intervention. The System Operates Effectively In Low-light And Crowded Indoor Conditions Through Adaptive Preprocessing, Requires No Wearable Devices, And Maintains A Structured Incident Log For Auditing And Analysis, Representing A Substantial Advancement Toward Improving Real-time Threat Detection And Operational Security In Modern Banking Environments. Read more Read less

Inductive Based Industrial Automation

Unanticipated Machinery Failures In Industrial Environments, Arising From Overheating, Excessive Vibration, Electrical Faults, Or Fire, Can Result In Severe Equipment Damage, Operational Downtime, And Safety Hazards. This Study Presents A Novel IoT-b... Unanticipated Machinery Failures In Industrial Environments, Arising From Overheating, Excessive Vibration, Electrical Faults, Or Fire, Can Result In Severe Equipment Damage, Operational Downtime, And Safety Hazards. This Study Presents A Novel IoT-based Hardware-interlocked Protection System Designed For Real-time Monitoring And Autonomous Fault Prevention In Industrial Machines. Central To The System Is The ESP32 Microcontroller, Selected For Its Reliable Real-time Performance And Integrated Wi-Fi Capabilities, Enabling Seamless Connectivity With IoT Platforms. A Comprehensive Network Of Sensors Including Temperature, Vibration, Fire, Voltage, And Current Detectors Continuously Assesses Critical Machine Parameters. Deviations Beyond Pre-established Safety Thresholds Automatically Activate Protective Responses, Such As Audible Alerts Via A Buzzer, Fault Notifications On An LCD Display, And Hardware Interlocking Through Relays That Immediately Interrupt Motor Supply. Simultaneously, The System Transmits Sensor Readings And Fault Occurrences To A Cloud-based Platform, Supporting Remote Monitoring, Fault Logging, And Analytical Evaluation For Maintenance Optimization. By Integrating Hardware Interlocking With IoT-driven Supervision, The Proposed Framework Reduces Dependence On Manual Observation, Ensures Rapid Response To Anomalies, And Enhances Overall Operational Safety. Its Cost-effectiveness, Adaptability, And Scalability Make It Suitable For Diverse Industrial Settings, Aligning With Smart Manufacturing And Industry 4.0 Principles. The Solution Not Only Improves Machine Reliability But Also Extends Equipment Lifespan Through Predictive Maintenance And Proactive Fault Management. This Work Demonstrates That Coupling IoT Intelligence With Tangible Interlocking Mechanisms Establishes A Robust, Proactive Strategy For Industrial Machine Protection Read more Read less

Neural 3D Avatar-Based Bidirectional Communication System For Deaf And Hearing Users

Deaf And Mutes Use Sign Language To Communicate, Yet Communication Between Signers And Non-signers Is Challenging As There Is A Lack Of Awareness. This Lays Obstacles In Such Spheres As Education, Medical Care, And Government Services. The Current Sy... Deaf And Mutes Use Sign Language To Communicate, Yet Communication Between Signers And Non-signers Is Challenging As There Is A Lack Of Awareness. This Lays Obstacles In Such Spheres As Education, Medical Care, And Government Services. The Current Systems Have Shortcomings Such As Limited Vocabulary, Inadequate Real Time Operation, Support Of The Indian Sign Language And No. Effective Two-way Communication. This Project Sug Gests An AI-based Communication To Deal With These Challenges. System That Facilitates A Smooth Interaction Between Sign-language And Nonsigners. The Model Is Based On Transformer To Identify Continuous. Facial Expressions, Body Movements, And Gestures Of Live Video Input, Providing Proper And Real-time Interpretation Of The Sign Language. It Also Incorporates Speech Recognition And Text To Speech Modules In Order To Translate. Translates Voice Text And Voice Even In A Noisy Setting. Another Component Of A Neural Avatar Is A Neural Sign Animation To. Visually Represent Communication. The System Is Adopted As Web-based Application, Which Is Accessible. And Ease Of Use. Overall, The Proposed System Enhances Inclusive Communication And Breaks The Barrier Between Deaf And The Hearing Communities. Read more Read less

Enhanced ODIRNet: Attention-Driven And Explainable Deep Neural Network For Robust Diabetic Retinopathy Detection

Diabetic Retinopathy (DR) Is A Significant Global Health Concern, Especially In Areas With High Rates Of Diabetes. Early Detection Via Automated Systems Can Reduce The Risk Of Blindness. This Paper Introduces ODIRNet, A Compact Deep Convolutional Neu... Diabetic Retinopathy (DR) Is A Significant Global Health Concern, Especially In Areas With High Rates Of Diabetes. Early Detection Via Automated Systems Can Reduce The Risk Of Blindness. This Paper Introduces ODIRNet, A Compact Deep Convolutional Neural Network That Efficiently Classifies Retinal Fundus Images. ODIRNet Is Developed From The Ground Up, Incorporating Advanced Feature Extraction Techniques Such As Blue-channel Emphasis And Attention Modules. The Model Was Trained And Tested On The Ocular Disease Intelligent Recognition (ODIR) Dataset, Which Contains 6392 Images. Results Show That ODIRNet Achieves An Accuracy Of 89.70%, Surpassing Models Like VGG16, ResNet50, And MobileNet. Furthermore, The Model Is Integrated Into A Web-based Platform For Real-time Diagnostic Screening, Making It A Practical And Accessible Solution For Clinics With Limited Resources. Read more Read less

A Study On Postnatal Depression Among Mothers Availing Health Services At Government Kasturba Gandhi Hospital, Triplicane, Chennai District, TamilNadu: A Cross-Sectional Study

Background: Postpartum Depression (PPD) Is A Common Mental Health Disorder Affecting Women Following Childbirth, With Notable Consequences For Both Mother And Child. It Is Characterized By Persistent Sadness, Emotional Instability, Fatigue, And I... Background: Postpartum Depression (PPD) Is A Common Mental Health Disorder Affecting Women Following Childbirth, With Notable Consequences For Both Mother And Child. It Is Characterized By Persistent Sadness, Emotional Instability, Fatigue, And Impaired Daily Functioning. Unlike The Transient “baby Blues,” PPD Is More Severe And Long-lasting, Requiring Timely Recognition And Intervention. The Condition May Begin During Pregnancy Or Within The First Year After Delivery, Depending On Varying Clinical Definitions. Early Identification And Appropriate Management Are Essential To Reduce Its Impact On Maternal Well-being, Infant Development, And Family Dynamics. The Postpartum Period Is A Critical Phase Marked By Significant Physical, Hormonal, And Psychological Changes In A Woman’s Life. While Many Women Adjust Well, A Considerable Proportion Experience Mental Health Challenges, Among Which Postpartum Depression Is One Of The Most Prevalent. Factors Such As Hormonal Fluctuations, Lack Of Social Support, Stress, And Prior Mental Health Issues Contribute To Its Development. Despite Its Frequency, PPD Often Remains Underdiagnosed Due To Stigma, Lack Of Awareness, And Limited Screening In Routine Care. Addressing Postpartum Depression Is Essential For Improving Maternal Health Outcomes And Ensuring Healthy Child Development. Objectives: To Determine The Prevalence Of Post-Natal Depression Among Mothers In The Outpatient Department Of Kasturba Gandhi Hospital, Triplicane, Chennai-02, Tamil Nadu Methods: A Cross-sectional Study Was Conducted Among 370 Mothers Availing Health Services At Kasturba Gandhi Hospital. After Getting Consent From Them, Questionnaire For Collecting Socio-demographic Details And A Validated Tamil Version (CMC Vellore) Of Edinburgh Postnatal Depression Scale Was Used To Screen For Postnatal Depression Among Mothers. Data Was Collected Using Excel And Analysed Using SPSS Version 16. Result: Out Of 372 Participants, 87 Were Found To Have Depression(23.4%). It Was Found That PPD Is Significantly Associated With: Stress, Mental Health Problems And Any Other Comorbidities. Conclusion: The Present Study Reveals A Considerable Prevalence Of Postnatal Depression Among The Study Participants, Emphasizing The Need For Timely Identification And Intervention. Read more Read less

Amitext: Emotionally Intelligent Message Rewriting Using Transformer Models And Reinforcement Learning

Amitext Is A Premium Message Rewriting Software That Will Enhance The Standards Of Online Communication By Transforming Emotionally Toned Or Offensive Text Messages To Friendly, Sympathetic And Constructive Messages In A Manner That Will Not Corrupt... Amitext Is A Premium Message Rewriting Software That Will Enhance The Standards Of Online Communication By Transforming Emotionally Toned Or Offensive Text Messages To Friendly, Sympathetic And Constructive Messages In A Manner That Will Not Corrupt The Original Content Of The Text. Negative Or Insensitive Messages In Online Communities, E.g. Customer Service Portals, Peer-support Forums, And Mental Health Forums, Are Likely To Be Misleading, Create Conflict And Cause Emotional Trauma. Moderation Systems And Rule-based Rewriting Systems Cannot Produce The Emotionally Subtlety And In Most Cases The Systems Produce Responses That Are Grammatically Correct But Tone-deaf. A Different Solution To This Weakness Is Provided By Amitext, Which Integrates Transformer-based Language Models, Sentiment Classification, And Reinforcement Learning (RL) To Analyze Tone, Intent, And Meaning Jointly And Then Rewrite. It Is Remarkable Due To The Most Innovative Feedback Loop Of Adaptive Rewriting, That Serves To Refine The Quality Of The Rewriting In A Continuous Fashion, Owing To The Multi-objective Rewards On The Necessity To Retain The Meaning, Enhance Politeness, And Match Sentiment. Amitext Is A Tool, Unlike The Fixed Filters Or Template-driven Paraphrasers; The More It Interacts With The Human, The More It Becomes More And More Conscious. Amitext Is An Effective And Scalable Empathy Awareness, Tone Fully Customizable And Adaptive Learning To Promote Healthier Communication And Reduce Digital Friction On The Online Platforms. Read more Read less

An AI Approach To Human Non-Movement Analysis

This Paper Presents An AI-based Human Non-Movement Detection System Designed For Real-time Monitoring And Safety Applications. Traditional Systems Mainly Focus On Motion Detection And Fail To Identify Prolonged Inactivity, Which May Indicate Critical... This Paper Presents An AI-based Human Non-Movement Detection System Designed For Real-time Monitoring And Safety Applications. Traditional Systems Mainly Focus On Motion Detection And Fail To Identify Prolonged Inactivity, Which May Indicate Critical Situations. The Proposed System Uses YOLO For Human Detection And OpenCV For Frame Processing. It Continuously Analyzes Movement And Detects Inactivity Based On Predefined Time Thresholds. When Abnormal Stillness Is Detected, The System Generates Alerts Through Sound And Telegram Notifications. The System Is Efficient, Automated, And Suitable For Healthcare, Surveillance, And Workplace Safety Applications. Read more Read less

AI-Driven Accident Analysis And Real-Time Emergency Alert System

Road Accidents Are A Major Cause Of Death And Serious Injuries Worldwide, Substantially Due To Delayed Discovery And Exigency Response. This Design Presents An AI- Grounded Real- Time Road Accident Detection, Beget Analysis, And Future Accident Preve... Road Accidents Are A Major Cause Of Death And Serious Injuries Worldwide, Substantially Due To Delayed Discovery And Exigency Response. This Design Presents An AI- Grounded Real- Time Road Accident Detection, Beget Analysis, And Future Accident Prevention System Using CCTV Surveillance And Deep Literacy Ways. The System Continuously Monitors Road Business And Automatically Detects Accidents In Real Time Using The YOLOv8 Model. Once An Accident Is Detected, The System Estimates Its Inflexibility And Incontinently Sends Cautions With Position Details To The Nearest Hospitals To Insure Quick Medical Backing. In Addition, The System Analyzes Possible Causes Of Accidents Similar Asover-speeding, Collisions, And Business Violations While Storing All Incident Data For Farther Analysis. The Collected Data Is Used To Identify Accident-prone Areas And Induce Intelligent Forestallment Suggestions To Ameliorate Road Safety. Eventually, This System Reduces Exigency Response Time, Supports Authorities, And Helps In Minimizing Road Accident Losses Through AI- Driven Robotization And Analysis. Read more Read less

AI-Based Supply Chain Risk Prediction System Using Machine Learning

Supply Chain Systems Are Highly Vulnerable To Disruptions Caused By Factors Such As Transportation Delays, Traffic Congestion, Weather Conditions, And Inventory Fluctuations. Traditional Risk Assessment Methods Are Largely Manual, Time-consuming, And... Supply Chain Systems Are Highly Vulnerable To Disruptions Caused By Factors Such As Transportation Delays, Traffic Congestion, Weather Conditions, And Inventory Fluctuations. Traditional Risk Assessment Methods Are Largely Manual, Time-consuming, And Often Fail To Provide Accurate And Timely Insights. This Project Presents An Intelligent Web-based System For Predicting Supply Chain Risk Using Machine Learning Techniques. The Proposed System Utilizes A Random Forest Classifier To Analyze Key Operational Parameters, Including Delay, Traffic, Weather, Inventory, Order Value, And Port Delay. Based On These Inputs, The System Classifies Supply Chain Risk Into Three Categories: Low, Medium, And High. The Application Is Developed Using The Flask Framework And Integrates A User-friendly Interface For Manual Data Entry, CSV-based Batch Prediction, And Real-time Analytics Through A Dashboard. The System Also Includes Data Storage Using SQLite And Visualization Features Such As Charts And Export Functionalities For Excel And Image Formats. Experimental Results Indicate Moderate Model Performance, With An Accuracy Of Approximately 61.9% And Balanced Accuracy Of Around 67%, Highlighting The Need For Further Optimization. Despite These Limitations, The System Demonstrates The Practical Application Of Machine Learning In Supply Chain Risk Prediction And Provides A Foundation For Future Enhancements In Predictive Logistics Systems. Read more Read less

A Deep Learning–Driven Intelligent Framework For Automated Prediction And Detection Of Autism And ADHD

Autism Spectrum Disorder (ASD) And Attention Deficit/Hyperactivity Disorder (ADHD) Are Neuro Developmental Conditions That Affect The Cognitive Behavior, Attention, And Social Interaction. Early And Accurate Identification Is Essential For A Effectiv... Autism Spectrum Disorder (ASD) And Attention Deficit/Hyperactivity Disorder (ADHD) Are Neuro Developmental Conditions That Affect The Cognitive Behavior, Attention, And Social Interaction. Early And Accurate Identification Is Essential For A Effective Intervention; However, Traditional Diagnostic Of A Approaches Often Rely On Subjective Clinical Assessments. This Paper Proposes A Computational Framework For The Automated Screening Of Autism And ADHD. The System Analyzes Facial Images For Autism Detection And MRI Brain Images For ADHD Detection Using Structured Analytical Methods. In Addition, Behavioral Screening Tests Have A Such As Camera-based Observation, Speech Pattern Analysis, And Questionnaire-based Evaluation Are Integrated To Support Clinical Assessment. The Framework Is Implemented As A Web-based Platform That Enables Data Upload, Automated Evaluation, And Report Generation. The Proposed System Aims To Assist Early Identification, Support Clinical Decision-making, And Improve Accessibility To Neuro Developmental Disorder Screening. Read more Read less

A CROSS SECTIONAL STUDY ON KNOWLEDGE REGARDING BREAST SELF EXAMINATION (BREAST SELF-EXAMINATION) AMONG WOMEN IN THE REPRODUCTIVE AGE GROUP (15 TO 45 YEARS) IN THE GOVERNMENT OMANDURAR MEDICAL COLLEGE AND HOSPITAL

ASSESSMENT AND REDUCTION OF CANAL WATER POLLUTION AT AQUACULTURE WASTEWATER TREATMENT

Aquaculture Wastewater Is A Major Source Of Pollution In Canal Water Because It Contains Organic Waste, Nutrients, Salts, And Harmful Microorganisms. This Study Looks At Pollution Levels And Ways To Lessen Its Impact Using Simple Treatment Methods. W... Aquaculture Wastewater Is A Major Source Of Pollution In Canal Water Because It Contains Organic Waste, Nutrients, Salts, And Harmful Microorganisms. This Study Looks At Pollution Levels And Ways To Lessen Its Impact Using Simple Treatment Methods. Water Samples Were Collected From The Aquaculture Pond, Nearby Areas, And The Canal. They Were Tested For PH, Turbidity, Salinity, And BOD. The Results Showed That The Water Is Highly Polluted And Not Suitable For Direct Irrigation. To Improve Water Quality, We Used Chemical Treatment With Boric Acid, Sodium Hydroxide, And Sodium Tetraborate, Along With Sand Filtration As A Physical Treatment Method. This Combined Treatment Effectively Reduced Pollution And Improved Water Quality. This Method Is Simple, Cost-effective, And Supports Sustainable Aquaculture And Environmental Protection. Read more Read less

Construction Management Of Data Center And Energy Infrastructure

The Rapid Digital Transformation And Exponential Growth In Data Consumption Have Driven An Unprecedented Surge In Data Center Construction Worldwide, Placing Immense Pressure On Energy Infrastructure And Raising Critical Concerns About Sustainability... The Rapid Digital Transformation And Exponential Growth In Data Consumption Have Driven An Unprecedented Surge In Data Center Construction Worldwide, Placing Immense Pressure On Energy Infrastructure And Raising Critical Concerns About Sustainability And Operational Efficiency. This Study Examines The Evolving Landscape Of Data Center And Energy Infrastructure Projects Through A Comprehensive Analysis Of Current Trends, Persistent Challenges, And Emerging Best Practices In Their Construction And Deployment. The Research First Analyzes Key Industry Trends, Including Hyperscale Facility Development, Edge Computing Expansion, And The Integration Of Advanced Cooling Systems, Alongside Major Challenges Such As Escalating Power Demands, Supply Chain Disruptions, Skilled Labor Shortages, And Regulatory Hurdles. It Further Evaluates Best Practices In Modular Construction, Prefabrication Techniques, And Risk Management Frameworks That Have Demonstrated Success In Accelerating Project Timelines And Reducing Costs. Read more Read less

Analysis Of Interface Risks And Coordination Challenges In Multi-Contractor Campus Development Projects

Multi-contractor Campus Development Projects Are Increasingly Adopted By Universities To Accelerate Delivery And Leverage Specialist Expertise, Yet They Consistently Suffer From Significant Interface Risks And Coordination Failures That Lead To Delay... Multi-contractor Campus Development Projects Are Increasingly Adopted By Universities To Accelerate Delivery And Leverage Specialist Expertise, Yet They Consistently Suffer From Significant Interface Risks And Coordination Failures That Lead To Delays, Cost Overruns, Rework, And Operational Disruption In Live Academic Environments. Despite Growing Literature On Interface Management In Megaprojects And Infrastructure, The Distinctive Combination Of Fragmented Contracting, Strict Phasing Requirements, Sustainability Mandates, And The Necessity To Maintain Uninterrupted Teaching And Research Activities On Campus Remains Largely Unaddressed. This Research Aims To Fill This Gap By Systematically Analysing Interface Risks And Coordination Challenges Specific To Multi-contractor University Campus Projects In Developed Economies. Using A Sequential Explanatory Mixed-methods Design, The Study Will First Identify And Prioritise Interface Risks Through A Large-scale International Questionnaire Survey Of Experienced Practitioners. Subsequently, In-depth Semi-structured Interviews, Participatory Workshops, And Multiple Case Studies Of Recent Campus Projects (2018–2025) Will Explore Root Causes, Interdependencies, And Campus-specific Aggravating Factors. The Quantified And Qualitative Findings Will Then Be Synthesised To Develop A Practical, Four-layer Interface Management Framework (Governance – Process – Tools – Behaviours) Tailored For Multi-prime Higher-education Projects. The Proposed Framework Will Be Refined And Validated Through A Two-round Delphi Process With An International Expert Panel. Read more Read less

Effectiveness Of Different Contract Types In Mitigating Owner And Contractor Risks

This Research Investigates How Four Principal Contract Types — Lump Sum, Item Rate, Cost-Plus, And EPC Turnkey — Distribute And Mitigate Risks Between Owners And Contractors Throughout The Entire Project Life Cycle, From Early Concept And Design Thro... This Research Investigates How Four Principal Contract Types — Lump Sum, Item Rate, Cost-Plus, And EPC Turnkey — Distribute And Mitigate Risks Between Owners And Contractors Throughout The Entire Project Life Cycle, From Early Concept And Design Through Procurement, Construction, Commissioning, Handover, And The Defects Liability Period. The Study Compares Their Real-world Performance In Delivering Cost And Schedule Certainty For Owners While Preserving Profit Stability And Financial Health For Contractors, Analyses The Influence Of Varying Project Characteristics On Each Contract Type’s Effectiveness, And Identifies The Specific Conditions Under Which Traditional Models Break Down And When Collaborative Or Incentive-based Approaches Produce Markedly Superior Outcomes. Read more Read less

ANALYSIS OF MOUTH SELF EXAMINATION AWARENESS AMONG MEDICAL STUDENTS

Introduction: Oral Cancer Remains A Major Public Health Concern In India, With High Morbidity And Mortality Largely Due To Late Diagnosis. Mouth Self-examination (MSE) Is A Simple, Cost-effective Method For Early Detection Of Potentially Malignant Di... Introduction: Oral Cancer Remains A Major Public Health Concern In India, With High Morbidity And Mortality Largely Due To Late Diagnosis. Mouth Self-examination (MSE) Is A Simple, Cost-effective Method For Early Detection Of Potentially Malignant Disorders. However, Its Awareness And Practice Among Future Healthcare Providers Remain Uncertain. Objectives:To Assess The Knowledge, Awareness, And Practice Of Mouth Self-examination Among Undergraduate MBBS Students. Methods:A Cross-sectional Study Was Conducted Among 150 Undergraduate MBBS Students (first To Third Year) At Government Medical College, Omandurar Government Estate, Chennai, Over Two Months (May–June 2026). Participants Were Selected Using A Calculated Sample Size Of 150. Data Were Collected Using A Pre-designed, Self-administered Structured Questionnaire Distributed Via Google Forms. The Questionnaire Assessed Socio-demographic Details And Domains Related To Knowledge, Awareness, And Practice Of MSE. Data Were Analyzed Using Statistical Package For Social Sciences Software ( Version 16 ). Results:Among The Participants, 50.7% Had Heard Of MSE, And 59.3% Correctly Identified Its Purpose. While 64.7% Were Aware Of The Components Of MSE And 68.7% Recognized Its Role In Early Detection Of Oral Cancer, Only 32% Had Ever Performed MSE. Knowledge Regarding Risk Factors Was High, With 84.7% Identifying Tobacco As A Major Risk Factor. However, Only 49.3% Could Differentiate Between White And Red Lesions. Despite Limited Practice, 86.7% Expressed Willingness To Recommend MSE To Others. Conclusion:The Study Reveals Moderate Awareness But Poor Practice Of Mouth Self-examination Among Medical Students, With Notable Gaps In Detailed Clinical Knowledge. Integrating Structured Training And Practical Demonstrations Into The Undergraduate Curriculum Is Essential To Bridge The Gap Between Knowledge And Practice, Thereby Enhancing Early Detection Of Oral Cancer. Read more Read less

AI-Based Financial Health & Investment System

The AI-Based Personal Finance Besides Investment Planning System Operates As A Working Prototype Designed To Enhance The Detection Of Individual Credit Risk And Support Long-term Solvency Through An Automated Financial Structure. While Traditional Cr... The AI-Based Personal Finance Besides Investment Planning System Operates As A Working Prototype Designed To Enhance The Detection Of Individual Credit Risk And Support Long-term Solvency Through An Automated Financial Structure. While Traditional Credit Models Often Fail To Measure How A Single Person Withstands Financial Stress, This System Addresses That Gap By Processing Raw User Data Including Income, Expenditures, Savings, And Risk Appetite. The Program Generates A Comprehensive Financial Health Score (0–100) And Calculates The Specific "FIRE" (Financial Independence, Retire Early) Target Sum Required For Early Retirement. It Further Categorizes User Pathssuch As Lean, Fat, Barista, Or Coastand Identifies Critical Gaps In Insurance Coverage And Emergency Fund Adequacy. By Utilizing A Transparent, Logic-driven Recommendation Engine, The Tool Provides Evidence-based Prompts That Curb Personal Credit Risk And Guide Users Toward Sustainable Wealth Growth. Read more Read less

An Eco-friendly Approach To Bioplastic Production Using Prosophis Julifloraa Strach

As Syenthetic Orpetroleum Based Plastics Create A Severe Environmental Impact, An Attempt Is Made To Develop Sustainable Bioplastic From The Invasive Plant Species. Starch Extract From The Prosopisjuliflorastarch. Aneco-friendly Bioplastics To Satisf... As Syenthetic Orpetroleum Based Plastics Create A Severe Environmental Impact, An Attempt Is Made To Develop Sustainable Bioplastic From The Invasive Plant Species. Starch Extract From The Prosopisjuliflorastarch. Aneco-friendly Bioplastics To Satisfy The Requirements Of Industrial And Commercial Applications, A Common Limitation Has Been Their Reliance On Edible Agricultural Resources Such As Corn, Potatoes, Rice And Some Other Food Grains.The Research Provides A Sustainable Solution By Extracting Starch From The Pods Of Prosopis Juliflora, A Plant Species Widely Recognized For Its Invasive Water Solubility And Absorption Test To Assess Hydro-instability, And A Water Contact Angle Test To Evaluate Surface Characteristics. Crucially, The Environmental Viability Of The Materials Was Determined Through A Rigorous Biodegradability Test And A Morphological Analysis Was Conducted To Examine Their Microstructural Integrity. The Findings From These Analyses Conclusively Reveal That The Bioplastics Developed From Prosopis Juliflora Starch Possess Properties That Position Them As A Superior Alternative To Conventional Plastics For Packaging Applications. The Adoption Of This Material Offers A Dual Environmental Benefit. The Result Reveals That The Starch Based Bioplastics Would Be Better Alternative Material To Be Used In Several Packaging Industries, Not Only Does It Provide A Sustainable And Biodegradable Substitute For Synthetic Polymers, But It Also Leverages A Problematic Invasive Species As A Valuable Raw Material, Thereby Contributing Significantly To Environmental Remediation And Resource Management. Read more Read less

INTERPRETABLE AI-BASED RESOURCE ALLOCATION FOR VIRTUAL MACHINES IN CLOUD PLATFORMS

Cloud Computing Environments Require Efficient Storage Allocation Mechanisms To Handle The Rapid Growth Of Data While Maintaining Performance And Scalability. This Research Emphasizes Intelligent Storage Management As A Core Component By Integrating... Cloud Computing Environments Require Efficient Storage Allocation Mechanisms To Handle The Rapid Growth Of Data While Maintaining Performance And Scalability. This Research Emphasizes Intelligent Storage Management As A Core Component By Integrating Virtualization With Dynamic Resource Allocation Strategies. Virtual Machines Are Utilized To Efficiently Distribute Storage And Computational Resources Across Physical Infrastructures, Ensuring Optimal Usage Of Available Capacity. Continuous Monitoring Of System Workloads Enables Adaptive Allocation Of Storage Resources, Reducing Redundancy And Preventing Inefficient Utilization. The Approach Enhances System Performance By Minimizing Storage Overhead And Ensuring Balanced Distribution Of Data Across The Cloud Environment. To Further Strengthen Storage Efficiency, A Time-To-Live (TTL) Based Data Self-destruction Mechanism Is Incorporated To Automatically Remove Expired Or Unnecessary Data. This Ensures That Storage Space Is Continuously Optimized Without Manual Intervention, Reducing Maintenance Complexity And Operational Costs. In Addition, Data Security Is Reinforced Through Blowfish Encryption Along With Secure Key Management Techniques Such As Key Rotation And Controlled Key Distribution. This Combination Of Intelligent Storage Allocation, Automated Data Lifecycle Management, And Strong Security Mechanisms Provides A Comprehensive Solution For Achieving Efficient, Secure, And Scalable Cloud Storage Systems. Read more Read less

Cyberbullying Detection And Prevention In Social Networks

The Rapid Growth Of Social Media Platforms Has Significantly Increased The Spread Of Harmful Online Content, Including Cyberbullying And Hate Speech. These Forms Of Communication Negatively Impact Individuals And Communities, Often Leading To Psychol... The Rapid Growth Of Social Media Platforms Has Significantly Increased The Spread Of Harmful Online Content, Including Cyberbullying And Hate Speech. These Forms Of Communication Negatively Impact Individuals And Communities, Often Leading To Psychological Distress And Social Conflicts. Existing Content Filtering Systems Are Limited In Their Ability To Effectively Detect And Prevent Such Behavior In Real Time. This Paper Proposes An Intelligent Cyberbullying Detection And Prevention System Using Natural Language Processing (NLP) And The VADER Sentiment Analysis Technique. The System Analyzes User-generated Content, Classifies Sentiments, And Filters Offensive Messages Based On Predefined Rules. A Blacklist Mechanism Is Implemented To Identify Repeat Offenders, While Real-time Alerts Notify Users And Administrators Of Harmful Activity. The Proposed System Enhances Online Safety By Providing An Adaptive, Efficient, And User-friendly Solution For Managing Social Media Interactions. Read more Read less

Real-Time Violence Detection Using Deep Learning

Violence Detection In Surveillance Videos Has Become An Essential Requirement For Modern Safety Systems Due To Rising Security Concerns Across Public, Private, And Institutional Environments. Traditional CCTV Monitoring Systems Rely Heavily On Manual... Violence Detection In Surveillance Videos Has Become An Essential Requirement For Modern Safety Systems Due To Rising Security Concerns Across Public, Private, And Institutional Environments. Traditional CCTV Monitoring Systems Rely Heavily On Manual Observation, Which Is Prone To Human Error, Delayed Response, And Inefficiency During High-risk Situations. Recent Advancements In Deep Learning And Computer Vision Have Enabled Intelligent Surveillance Systems Capable Of Identifying Violent Activities Automatically And In Real Time. This Paper Presents A Deep-learning-based Framework For Real-time Violence Detection Using MobileNetV2 And OpenCV. Transfer Learning, Data Augmentation, And Fine- Tuning Techniques Were Used To Enhance Model Accuracy And Address Data Imbalance. A Prediction-smoothing Algorithm Based On Moving Average Improves Detection Stability By Minimizing Frame-level Noise. The System Triggers An Audio Alert During Violence Events To Enable Immediate Intervention. Experimental Results Demonstrate That The Proposed Model Achieves High Accuracy, Stable Real-time Performance, And Successful Detection In Live Video Streams. This Solution Can Be Deployed In Institutions, Smart Cities, Public Spaces, And Home Monitoring Applications. Read more Read less

HIGH ACCURACY LIGHTWEIGHT IMAGE CLASSIFICATION USING AN IMPROVED YOLO AND VGG16

Waste Classification Plays A Crucial Role In Sustainable Waste Management By Categorizing Materials Based On Their Type To Ensure Proper Disposal And Recycling. Traditional Waste Sorting Methods, Which Rely Heavily On Manual Labor, Are Time-consuming... Waste Classification Plays A Crucial Role In Sustainable Waste Management By Categorizing Materials Based On Their Type To Ensure Proper Disposal And Recycling. Traditional Waste Sorting Methods, Which Rely Heavily On Manual Labor, Are Time-consuming, Error-prone, And Inefficient At Scale. The Exponential Increase In Global Waste Production Necessitates More Accurate And Automated Solutions. This Paper Proposes A Smart Waste Classification System That Integrates Two Deep Learning Paradigms: The VGG16 Convolutional Neural Network (CNN) For High-accuracy Feature-based Classification And The YOLO (You Only Look Once) Framework For Real-time Object Detection. Waste Images Are Preprocessed Through Normalization, Noise Filtering, And Data Augmentation Before Being Fed Into The Dual-model Pipeline. The VGG16 Model, Leveraging Transfer Learning From ImageNet Weights, Classifies Waste Into Six Categories Cardboard, Glass, Metal, Paper, Plastic, And Trash With High Precision. Concurrently, YOLO Identifies And Localizes Waste Items In Live Camera Feeds, Enabling Real-time Sorting Decisions Communicated Via Email And SMS Alerts. Experimental Results Demonstrate That The Integrated System Achieves Superior Classification Accuracy And Processing Speed Compared To Existing Single-model Approaches, Making It A Viable Solution For Automated Waste Management In Smart Cities, Recycling Facilities, And Industrial Environments. Read more Read less

COMPARATIVE STUDY OF RCC & PRESTRESSED RETAINING WALL

Retaining Walls Are Critical Structures Used To Resist Lateral Earth Pressure And Maintain Ground Stability In Civil Engineering Projects. Traditionally, Reinforced Cement Concrete (RCC) Retaining Walls Have Been Widely Used Due To Their Simplicity A... Retaining Walls Are Critical Structures Used To Resist Lateral Earth Pressure And Maintain Ground Stability In Civil Engineering Projects. Traditionally, Reinforced Cement Concrete (RCC) Retaining Walls Have Been Widely Used Due To Their Simplicity And Reliability. However, Prestressed Retaining Walls Have Emerged As A Modern Alternative Offering Improved Structural Efficiency. This Paper Presents A Comparative Study Between RCC And Prestressed Retaining Walls Based On Parameters Such As Structural Performance, Economy, Material Efficiency, Construction Time, And Durability. The Study Concludes That Prestressed Retaining Walls Provide Superior Performance In Terms Of Reduced Deflection, Improved Load-carrying Capacity, And Material Savings, Though Initial Costs And Complexity Are Higher Read more Read less

Smart AI – Powered Health Care System

- This Paper Presents A Smart AI-Based Healthcare System Designed To Enhance Disease Prediction And Support Clinical Decision-making Using Machine Learning Techniques. The System Analyzes Patient Data Such As Symptoms, Medical History, And Clinical... - This Paper Presents A Smart AI-Based Healthcare System Designed To Enhance Disease Prediction And Support Clinical Decision-making Using Machine Learning Techniques. The System Analyzes Patient Data Such As Symptoms, Medical History, And Clinical Records To Identify Patterns And Predict Diseases At An Early Stage. It Improves Accuracy, Reduces Processing Time, And Assists Healthcare Professionals. Read more Read less

Heart Disease Prediction System

Heart Disease Is One Of The Leading Causes Of Mortality Worldwide, Making Early Risk Prediction Essential For Preventive Healthcare. Machine Learning (ML) Has Become An Effective Tool For Identifying Hidden Patterns In Medical Data To Support Clinica... Heart Disease Is One Of The Leading Causes Of Mortality Worldwide, Making Early Risk Prediction Essential For Preventive Healthcare. Machine Learning (ML) Has Become An Effective Tool For Identifying Hidden Patterns In Medical Data To Support Clinical Decision-making. This Research Focuses On Developing A Heart Disease Prediction System Using Machine Learning Algorithms Such As Logistic Regression, Random Forest, Support Vector Machine (SVM), And K-Nearest Neighbors (KNN). The Dataset Includes Key Clinical Parameters Such As Age, Cholesterol Levels, Blood Pressure, Chest Pain Type, And ECG Results. Performance Evaluation Is Carried Out Using Accuracy, Precision, Recall, And F1-score. Among The Models Tested, The Random Forest Classifier Achieved The Highest Accuracy, Demonstrating That ML-based Systems Can Significantly Improve Early Detection And Reduce Risk Through Timely MedicalHeart Disease Remains One Of The Most Life-threatening And Widely Spread Medical Conditions Across The Globe, Contributing To A Significant Percentage Of Deaths Each Year. Early Identification Of Individuals Who Are At High Risk Can Drastically Reduce Mortality Through Timely Treatment And Lifestyle Modifications. However, Conventional Diagnostic Methods Depend Heavily On Clinical Expertise And Manual Interpretation, Which May Lead To Inconsistent Results. In Recent Years, Machine Learning (ML) Has Emerged As A Powerful Analytical Technique Capable Of Learning Patterns From Medical Data And Providing Reliable Predictive Insights.This Study Aims To Design And Evaluate An Intelligent Heart Disease Prediction Model Using Multiple Machine Learning Algorithms, Including Logistic Regression, Support Vector Machine (SVM), Random Forest, And K-Nearest Neighbors (KNN). The System Analyzes Crucial Health Indicators Such As Age, Blood Pressure, Cholesterol Level, Chest Pain Type, Blood Sugar, And ECG Readings To Determine The Probability Of Heart Disease In A Patient. Performance Metrics Such As Accuracy, Sensitivity, Specificity, Precision, And F1-score Are Utilized To Identify The Best-performing Algorithm Intervention. Read more Read less

Wifi Deauthendication Detector

This Paper Presents A Low-cost Wi-Fi Security Monitoring System Designed To Detect Deauthentication Attacks In Real Time Using The ESP8266 NodeMCU Microcontroller. The Increasing Reliance On Wireless Networks Has Made Network Security An Urgent Prior... This Paper Presents A Low-cost Wi-Fi Security Monitoring System Designed To Detect Deauthentication Attacks In Real Time Using The ESP8266 NodeMCU Microcontroller. The Increasing Reliance On Wireless Networks Has Made Network Security An Urgent Priority. Deauthentication Attacks Exploit Unprotected IEEE 802.11 Management Frames To Disconnect Devices Without Authorization. The Proposed System Operates In Promiscuous Mode, Enabling It To Capture And Analyze All IEEE 802.11 Management Frames Within Radio Range. The Firmware Identifies Malicious Deauthentication (0xA0) And Disassociation (0xC0) Frame Types Commonly Exploited In Denial-of-service (DoS) Attacks. Upon Detection, The System Triggers Multi-modal Alerts Including LED Indication, Piezoelectric Buzzer Alarm, And OLED Display Notifications. Experimental Results Confirm 24x7 Monitoring Across All Wi-Fi Channels (1 Through 13) With Sub-300 Ms Detection Latency And Zero False Positives In A Realistic Office Environment Containing 14 Active Access Points. Read more Read less

Secure And Scalable Restful Banking System Using Spring Boot Framework

This Project Shows How To Design And Build A Secure, High-performance Online Banking REST API With Spring Boot 3.x. The System Uses A Microservices Architecture To Make It Easier To Scale And Maintain. OAuth 2.0, OpenID Connect, And JWT-based Authent... This Project Shows How To Design And Build A Secure, High-performance Online Banking REST API With Spring Boot 3.x. The System Uses A Microservices Architecture To Make It Easier To Scale And Maintain. OAuth 2.0, OpenID Connect, And JWT-based Authentication With RBAC Protect Important Features Like Managing Accounts, Transferring Money, And Checking Balances. Transaction Management That Follows ACID Standards Makes Sure That All Financial Operations Are 100% Consistent And Reliable. Performance Testing Shows That The Average Response Time Is Less Than 30 Ms And The Throughput Is More Than 2000 TPS When The Load Is High. Security Features Like Mutual TLS And Fraud Detection Patterns Cut Down On Vulnerability Exploits By 84%. The System Is A Scalable, Secure, And Cloud-ready Solution For Modern Digital Banking Platforms. Read more Read less

AI-Driven Resume Analysis And Job-Specific Resume Optimization System

This Paper Presents An AI-driven System For Analyzing Resumes And Optimizing Them To Match Specific Job Descriptions. Modern Recruitment Relies Heavily On Applicant Tracking Systems (ATS), Which Filter Candidates Based On Keyword Relevance And Semant... This Paper Presents An AI-driven System For Analyzing Resumes And Optimizing Them To Match Specific Job Descriptions. Modern Recruitment Relies Heavily On Applicant Tracking Systems (ATS), Which Filter Candidates Based On Keyword Relevance And Semantic Alignment. Many Qualified Candidates Are Inadvertently Eliminated Due To Poor Resume Formatting Or Misaligned Content. The Proposed System Employs Natural Language Processing (NLP) And Machine Learning (ML) Techniques, Including Named Entity Recognition (NER) For Structured Information Extraction And Sentence-BERT (SBERT) Embeddings With Cosine Similarity For Semantic Matching Between Resumes And Job Descriptions. The System Identifies Skill Gaps, Computes A Quantitative Match Score, And Provides Personalized Optimization Recommendations Including Keyword Suggestions, Content Rephrasing, And ATS Formatting Guidance. A Flask-based REST API Enables Real-time Interaction. Experimental Evaluation Demonstrated An 88.40% Matching Accuracy, With 85% Of Users Reporting Improved Resume Quality. The System Offers An Accessible, Intelligent, And Explainable Solution To Bridge The Gap Between Job Seekers And Automated Recruitment Processes. Read more Read less

Virtual Queueing System For RationShop

Ration Shop's Virtual Queueing System Is An Online Program That Attempts To Modernize And Improve The Way People Access Food And Goods From Government Ration Shops. Many People Are Frustrated With Having To Wait Long Periods Of Time In Crowded Ration... Ration Shop's Virtual Queueing System Is An Online Program That Attempts To Modernize And Improve The Way People Access Food And Goods From Government Ration Shops. Many People Are Frustrated With Having To Wait Long Periods Of Time In Crowded Ration Stores Because There Were Not Enough People On Staff, There Were No Accurate Records Kept As To How Much Product Was Available For Purchase, And Employees Were Not Adequately Trained On The Procedures For Distributing The Products To The Needy. By Providing An Internet-based Application For The Distribution Of Rations, The Virtual Queueing System Connects Shopkeepers With Users With The Enablement Of An Administrator In The Coordinating Effort To Provide Improved Service Quality. The Administrator Of The Application Is Responsible For Registering The Shopkeepers, Approving Requests From Users, Maintaining A Record Of Each Shopkeeper's Employee Leave And Managing Each User's Ration Card. The Shopkeeper Can Provide Their Availability Of Product By Entering This Information In Real-time, Reserving Tokens For The User And Updating Users On The Reservation Status. Users May View The Shopkeeper's Available Quantity Of Product, The Shopkeeper's Leave Of Absence Status, Reserve A Token To Receive The Ration And Track The Status Of Their Token Reservation. Users May Also Apply And Request For New Ration Cards Or The Addition/removal Of Family Members, Which The Admin Reviews And Approves. The Admin Will Then Upload The New Ration Card For The User To Securely Download Via The System. These Processes Have Been Digitized Thereby Reducing The Amount Of Manual Work, Reducing Waiting Time, And Preventing Congestion At Ration Shops. The Digitization Of These Processes Has Also Improved Transparency, Accountability, And Fairness In Ration Distribution. Overall, The Virtual Queueing System For Ration Shops Is A User-friendly, Efficient, And Reliable Option That Improves Public Service Delivery And Provides Users Greater Access To Vital Commodities. Read more Read less

Ensemble Machine Learning For NIFTY-50 Price Forecasting And Trend Classification: A Flask-Deployed Decision Support System

Stock Market Forecasting Poses A Significant Challenge Due To The Non-linear, High-volatility Nature Of Financial Time Series. This Paper Presents An End-to-end Machine Learning Pipeline For Predicting NIFTY-50 Closing Prices And Next-day Directional... Stock Market Forecasting Poses A Significant Challenge Due To The Non-linear, High-volatility Nature Of Financial Time Series. This Paper Presents An End-to-end Machine Learning Pipeline For Predicting NIFTY-50 Closing Prices And Next-day Directional Trends. The System Trains Random Forest (RF) And Decision Tree (DT) Regressors On Historical OHLCV Data Augmented With Engineered Technical Features (MA10, MA50, Daily Returns). A Fusion Mechanism Averages RF And DT Outputs To Produce A Stabilized Price Estimate. A Separate RF Classifier Outputs Categorical Trend Labels (UP/DOWN/NEUTRAL), Avoiding The Pitfall Of Inferring Direction From Regression Residuals. Experimental Results Show That The RF+DT Fusion Achieves An R² Of 0.9451, Outperforming Standalone RF (0.9312) And DT (0.8841) Regressors. The Trend Classifier Achieves 82.4% Accuracy And An F1-score Of 0.81. The Complete Pipeline Is Deployed As A Flask Web Application Supporting User Authentication, Interactive Prediction, Candlestick Visualization, CSV Upload, Live Data Fetch Via Yahoo Finance, PDF Report Export, And An Administrative Panel. The System Provides A Practical, Interpretable, And Deployable Solution For Short-term NIFTY-50 Decision Support. Read more Read less

Dynamic Keystroke And Biometric User Authentication

Traditional Authentication Mechanisms Such As Passwords And PINs Are Widely Used For System Security; However, They Remain Vulnerable To Attacks Such As Password Theft, Shoulder Surfing, And Replay Attacks. To Address These Limitations, This Research... Traditional Authentication Mechanisms Such As Passwords And PINs Are Widely Used For System Security; However, They Remain Vulnerable To Attacks Such As Password Theft, Shoulder Surfing, And Replay Attacks. To Address These Limitations, This Research Proposes A Dynamic Multi-modal Biometric Authentication System That Integrates Keystroke Dynamics And Face Recognition Using A Support Vector Machine (SVM) Classifier. The Proposed System Continuously Verifies User Identity By Analyzing Typing Behaviour And Facial Features Captured Through A Webcam. Keystroke Timing Characteristics Such As Dwell Time And Flight Time Are Extracted, While Facial Features Are Obtained Using Computer Vision-based Face Recognition Techniques. The Proposed Model Demonstrates Strong Potential For Applications Requiring Continuous And Secure Authentication Such As Online Banking And Financial Platforms. Read more Read less

Real-Time Weather Monitoring Dashboard Using API

Weather Monitoring Has Become A Critical Aspect Of Everyday Decision-making For Individuals, Industries, And Government Agencies. Traditional Weather Reporting Systems Often Suffer From Delayed Data Updates, Limited Location-specific Information, And... Weather Monitoring Has Become A Critical Aspect Of Everyday Decision-making For Individuals, Industries, And Government Agencies. Traditional Weather Reporting Systems Often Suffer From Delayed Data Updates, Limited Location-specific Information, And Poor User Interfaces. In This Paper, We Present A Real-Time Weather Monitoring Dashboard, A Web-based Application That Integrates Third-party Weather APIs To Retrieve, Process, And Visualize Live Meteorological Data Including Temperature, Humidity, Wind Speed, Atmospheric Pressure, And Weather Conditions. The System Provides A Responsive And Interactive User Interface That Delivers Accurate, Up-to-date Weather Information For Any Searched Location Across The Globe. By Combining RESTful API Integration, Dynamic Data Visualization, And A Clean Front-end Design, The Proposed Dashboard Improves Accessibility And Usability For General Users, Farmers, Travelers, And Other Stakeholders Who Rely On Timely Weather Data For Planning And Decision-making. Read more Read less

AN AI-BASED HUMAN SCENE UNDERSTANDING MECHANISM FOR IMAGE CAPTIONING IN BLIND NAVIGATION AND ASSISTANCE

Visual Impairment Greatly Affects The Navigation And Interaction With The Surrounding Environment. Conventional Assistive Tools Are Generally Unintelligent, While Existing Digital Tools Are Generally Task-specific, Computationally Inefficient, And Re... Visual Impairment Greatly Affects The Navigation And Interaction With The Surrounding Environment. Conventional Assistive Tools Are Generally Unintelligent, While Existing Digital Tools Are Generally Task-specific, Computationally Inefficient, And Require Connectivity, Which Makes Them Less Effective In Real-time Applications. In This Paper, An Integrated AI-based Assistive Vision System Is Proposed, Which Can Improve The Situational Awareness Of The Visually Impaired Through The Use Of Multimodal Perception. The Proposed System Incorporates Object Detection, Face Recognition, And Currency Recognition In An Integrated Manner. Real-time Images Are Captured Using A Camera Module, Which Are Fed Into The Optimized Deep Learning Models For Image Processing. The YOLO Algorithm Is Used For Efficient Obstacle Detection, Allowing The Identification Of Moving Objects In The Surrounding Environment. Face Recognition Is Achieved Using The Grassmann Approach, Allowing The Robustness Of The Algorithm To Pose And Illumination Variations. A Convolutional Neural Network Is Used For Reliable Currency Classification, Even In Cases Of Partial Occlusion. The Processed Information Is Converted Into Context-aware Audio Feedback, Enabling Effective Navigation And Interaction Without Visual Dependency. Offline Functionality Ensures Operational Reliability In Diverse Conditions. The Integrated Architecture Enhances Accuracy, Reduces System Complexity, And Improves User Independence, Representing A Significant Advancement In Intelligent Assistive Technology. Read more Read less

REAL TIME CRASH DETECTION AND EMERGENCY ALERT SYSTEM

In Modern Traffic Conditions, Ensuring Road Safety Has Become Increasingly Important, As Timely Action During Accidents Can Significantly Reduce Fatalities And Injuries. This Project Presents An Intelligent Accident Detection And Notification System... In Modern Traffic Conditions, Ensuring Road Safety Has Become Increasingly Important, As Timely Action During Accidents Can Significantly Reduce Fatalities And Injuries. This Project Presents An Intelligent Accident Detection And Notification System That Combines Multiple Technologies Such As Arduino, An IoT Gateway, GPS, And MEMS-based Sensors. The System Relies On An ADXL335 Accelerometer To Continuously Track Sudden Variations In Motion, Which May Signal A Collision. When Such An Abnormal Movement Is Detected, The System Automatically Triggers An Alert And Transmits Key Information, Including The Exact Location, Through A Cloud-based IoT Platform To Preassigned Contacts. The Arduino UNO R3 Functions As The Central Controller, Managing Communication Between All Hardware Components Efficiently. A 16x2 LCD Screen Is Also Included To Display Real-time Information Such As Coordinates And System Status. This Solution Is Designed To Improve Emergency Response Efficiency By Enabling Quick And Accurate Accident Reporting, Ultimately Contributing To Safer Road Environments. Read more Read less

LRFS : Online Shopper's Behaviour - Based Efficient Customer Segmentation Model

Online Shopping Has Become A Dominant Platform In Modern Digital Commerce. Understanding Customer Behavior Is Essential For Improving Marketing Strategies And Increasing Revenue. This Paper Proposes An Enhanced Customer Segmentation Model Named LRFS,... Online Shopping Has Become A Dominant Platform In Modern Digital Commerce. Understanding Customer Behavior Is Essential For Improving Marketing Strategies And Increasing Revenue. This Paper Proposes An Enhanced Customer Segmentation Model Named LRFS, Which Extends The Traditional LRF Framework By Introducing A New Component Called Staying Rate. The Staying Rate Represents The Relationship Between User Engagement And Revenue Generation. The Model Utilizes Unsupervised Machine Learning Techniques Including K-Means And K-Medoids Clustering Along With Dimensionality Reduction Methods Such As PCA, T-SNE, And Autoencoder. Comparative Analysis Is Performed Between LR, LF, LRF, And The Proposed LRFS Model. Experimental Results Show That LRFS Provides More Accurate And Meaningful Customer Segmentation, Helping Businesses Identify High-value Customers And Improve Decision-making Strategies. Read more Read less

SMART BATTERY MONITORING AND PROTECTION SYSTEM FOR ELECTRIC VEHICLES USING IOT

Electric Vehicles (EVs) Require Efficient Battery Management To Ensure Safety, Reliability, And Longevity. This Project Proposes A Smart Battery Monitoring And Protection System That Continuously Tracks Critical Battery Parameters Such As Temperature... Electric Vehicles (EVs) Require Efficient Battery Management To Ensure Safety, Reliability, And Longevity. This Project Proposes A Smart Battery Monitoring And Protection System That Continuously Tracks Critical Battery Parameters Such As Temperature, Voltage, Current, Humidity, And Fire Hazards Using Sensors Integrated With An Arduino Uno. If Any Parameter Exceeds Safe Thresholds, The System Triggers A Relay To Disconnect The EV Motor. The IoT Module Enables Real-time Monitoring And Alerts, Allowing Users To Track Battery Status Remotely Through An Online Platform. This System Enhances EV Safety By Preventing Hazardous Conditions, Improving Operational Efficiency, And Extending Battery Lifespan. By Incorporating Automated Protection Mechanisms And IoT-based Monitoring, This Solution Provides A Comprehensive And Smart Approach To Battery Management In Electric Vehicles. The Project Aims To Contribute To The Advancement Of EV Technology By Offering A Cost-effective, Scalable, And Reliable Battery Safety Solution. Read more Read less

Smart Foot Step Power Generation

Smart Street Lighting Aims To Improve The Efficiency Of Street Lights By Automating Their Control As And When Needed, Without The Need For An External Power Source. Vehicles Moving Down The Road Cause Vibrations In The Piezoelectric Material Beneath... Smart Street Lighting Aims To Improve The Efficiency Of Street Lights By Automating Their Control As And When Needed, Without The Need For An External Power Source. Vehicles Moving Down The Road Cause Vibrations In The Piezoelectric Material Beneath The Road Due To Deformation Caused By Vehicle Passing Pressure. Piezoelectricity Is An Electric Charge That Builds Up In Certain Solid Materials (such As Crystals And Certain Ceramic Materials) When Torque Is Applied. Nowadays, Most Of The Existing Street Light Systems Are Wired Which Are Not Only Difficult To Construct But Also Have Poor Flexibility. To Overcome IOT Technology Using A Piezoelectric Sensor Which Uses Power Efficiently By Remotely Monitoring And Controlling The System. This Design Can Save A Better Amount Of Electricity Compared To Street Lamps That Keep A Light During The Night For This System Power Supply Is Given By A Solar Panel And Piezoelectric Transducer Placed On The Road. When A Vehicle Is Passing On-road Then Pressure Is Applied, According To The Property Of Piezoelectric Material Energy Is Generated When Pressure Is Applied To It. LDR Is A Minor Component That Works On The Intensity Of Light. When The Light Intensity Is Increased, LDR Resistance Increases, And When Light Intensity Is Decreased, LDR Resistance Decreases. The System Only Turns On Street Lights When They Are Needed To Maintain A Secure And Safe Traffic Environment. Light Automatically Dims Up Or Down In Partial Road Areas When The Street Lighting System Is Computer-controlled. Read more Read less

SURVEY ON SECURE MULTI-FACTOR GESTURE-ENHANCED BLOCKCHAIN E- VOTING SYSTEM

Electronic Voting Systems Are Increasingly Adopted To Improve Efficiency And Accessibility In Modern Democratic Processes. However, Traditional And Centralized Voting Systems Suffer From Issues Such As Voter Impersonation, Lack Of Transparency, Centr... Electronic Voting Systems Are Increasingly Adopted To Improve Efficiency And Accessibility In Modern Democratic Processes. However, Traditional And Centralized Voting Systems Suffer From Issues Such As Voter Impersonation, Lack Of Transparency, Centralized Control, And Vulnerability To Tampering. These Challenges Reduce Public Trust And Limit The Effectiveness Of Digital Voting Solutions. This Paper Proposes A Secure Multi- Factor Blockchain-based E-voting System Integrating Facial Recognition And Gesture- Based Interaction To Enhance Both Security And Usability. The System Employs Biometric Authentication To Verify Voter Identity And Uses Gesture Interaction As An Additional Confirmation Mechanism To Prevent Accidental Or Fraudulent Voting. Votes Are Encrypted Using SHA-256 Hashing And Stored In A Permissioned Blockchain Network Using Hyper Ledger Fabric, Ensuring Immutability, Transparency, And Auditability. The Proposed System Achieves Approximately 95% Authentication Accuracy, Prevents Duplicate Voting, And Ensures Voter Anonymity Through Identity– Vote Separation. Experimental Results Demonstrate That The System Improves Security, Accessibility, And Reliability Compared To Traditional And Centralized Voting Approaches. Read more Read less

Legal Judgment Prediction System Using Machine Learning

This Research Describes The Design And Implementation Of A Machine Learning-based Legal Judgment Prediction System That Includes Past Case Retrieval And Legal Reasoning Methods. Traditional Legal Research Relies On Manual Case Analysis And Keyword Se... This Research Describes The Design And Implementation Of A Machine Learning-based Legal Judgment Prediction System That Includes Past Case Retrieval And Legal Reasoning Methods. Traditional Legal Research Relies On Manual Case Analysis And Keyword Searches. These Methods Often Overlook Important Legal Context And Judicial Reasoning Patterns, Making It Difficult To Predict Case Outcomes.The Proposed System Offers A Smart Legal Analytics Framework. It Processes Legal Documents, Extracts Contextual Embeddings Using Legal Longformer, Predicts Case Outcomes With Supervised Learning Models, And Retrieves Past Judgments That Are Similar In Meaning. Legal Longformer Is Designed To Handle Long Legal Documents And Capture Long-range Contextual Relationships Across Judicial Texts. A Retrieval Engine Based On Semantic Similarity And A Reasoning Module Using A Retrieval-Augmented Generation (RAG) Approach Provide Clear And Understandable Decision Support.The System Follows A Modular Layered Architecture Integrating Preprocessing, Embedding Generation, Classification, Similarity Computation, And Web-based Deployment. Experimental Evaluation Demonstrates Reliable Prediction Accuracy, Effective Retrieval Of Relevant Precedents, And Improved Interpretability Through Reasoning Explanations. The Framework Improves Legal Research Efficiency By Combining Prediction, Retrieval, And Reasoning Into A Unified AI-driven Legal Decision Support System. Read more Read less

Dual Authentication Door Lock System Using ESP32

This Paper Presents A Dual Authentication Door Lock System Based On The ESP32 Microcontroller Aimed At Enhancing Security In Modern Access Control Applications. The System Employs Two Levels Of Authentication, Such As RFID, Password, Or Biometric Ver... This Paper Presents A Dual Authentication Door Lock System Based On The ESP32 Microcontroller Aimed At Enhancing Security In Modern Access Control Applications. The System Employs Two Levels Of Authentication, Such As RFID, Password, Or Biometric Verification, To Ensure That Only Authorized Users Can Gain Access. By Integrating Internet Of Things (IoT) Technology, The System Supports Remote Monitoring, Real-time Alerts, And Data Logging For Improved User Control And Convenience. The ESP32 Platform Is Utilized For Its Low Power Consumption, Built-in Wi-Fi And Bluetooth Capabilities, And Efficient Processing Performance. The Proposed System Offers A Reliable, Cost-effective, And Scalable Solution Suitable For Smart Homes And Industrial Security Applications. Read more Read less

AN IOT-ENABLED SMART WOMEN SAFETY AND EMERGENCY RESPONSE SYSTEM

Women’s Safety Has Become A Major Social And Technological Concern Due To The Increasing Number Of Harassment, Assault, And Emergency Incidents In Public And Private Spaces. Traditional Safety Mechanisms Such As Helplines And Mobile Applications Ofte... Women’s Safety Has Become A Major Social And Technological Concern Due To The Increasing Number Of Harassment, Assault, And Emergency Incidents In Public And Private Spaces. Traditional Safety Mechanisms Such As Helplines And Mobile Applications Often Require Manual Activation, Which May Not Be Possible During Critical Situations. The Internet Of Things (IoT) Offers An Effective Solution By Enabling Smart, Connected Devices That Can Automatically Sense Danger, Track Location, And Send Emergency Alerts In Real Time. By Integrating Sensors, Communication Modules, And Intelligent Decision Logic, An IoT-enabled Women Safety System Can Significantly Reduce Response Time And Provide Timely Assistance. This Project Proposes An IoT-based Smart Women Safety And Emergency Response System That Ensures Automatic Detection Of Distress Conditions And Instant Communication With Guardians And Emergency Services Read more Read less

Autonomous Human-Following Robot

Autonomous Human-following Robots Represent An Emerging Class Of Assistive Robotic Systems Aimed At Reducing Manual Effort In Transportation Tasks Across Environments Such As Healthcare, Logistics, And Personal Assistance. This Paper Presents The Des... Autonomous Human-following Robots Represent An Emerging Class Of Assistive Robotic Systems Aimed At Reducing Manual Effort In Transportation Tasks Across Environments Such As Healthcare, Logistics, And Personal Assistance. This Paper Presents The Design And Implementation Of A Cost-effective Autonomous Human-Following Transport Robot Based On An Arduino UNO Microcontroller Platform. The System Integrates Ultrasonic And Infrared Sensing Modules To Enable Real-time Distance Measurement And Obstacle Detection, While A Motor Driver Interface Ensures Controlled Actuation Of High-torque DC Gear Motors Mounted On A Four-wheel Drive Chassis. The Proposed Approach Focuses On Maintaining A Consistent And Safe Distance From A Human Target Using Sensor-based Feedback Control, While Simultaneously Ensuring Collision-free Navigation. The Control Logic Is Developed Using Embedded Programming Techniques, Enabling The Robot To Respond Dynamically To Environmental Changes. The Design Is Guided By Established Principles Of Mobile Robotics, Control Systems, And Embedded System Integration, As Discussed In Standard Literature And Technical Documentation. Experimental Evaluation Is Conducted In Structured Indoor Environments To Assess System Performance In Terms Of Tracking Accuracy, Response Time, And Obstacle Avoidance Capability. The Results Indicate That The System Achieves Reliable Human-following Behavior With Satisfactory Stability And Minimal Error Under Controlled Conditions. Although The Implementation Relies On Low-cost Sensors And Simplified Algorithms, It Demonstrates The Feasibility Of Developing Practical Assistive Robots Using Accessible Technologies. The Work Also Highlights Potential Extensions Involving Advanced Sensing And Intelligent Algorithms To Enhance Adaptability In Complex Real-world Scenarios. Read more Read less

AI-Driven Holographic Code Visualization And Collaborative Debugging Platform

This Paper Presents A Novel AI-driven Holographic Code Visualization And Collaborative Debugging Platform That Transforms Traditional Software Development Workflows. By Leveraging Advanced Artificial Intelligence Algorithms, Holographic Projection Te... This Paper Presents A Novel AI-driven Holographic Code Visualization And Collaborative Debugging Platform That Transforms Traditional Software Development Workflows. By Leveraging Advanced Artificial Intelligence Algorithms, Holographic Projection Technology, And Real-time Collaboration Frameworks, Our Platform Enables Developers To Visualize Code Structures, Execution Flows, And Debugging Information In Three-dimensional Holographic Space. The System Incorporates Natural Language Processing For Intelligent Code Analysis, Machine Learning For Predictive Debugging, And Augmented Reality Interfaces For Immersive Collaboration. Experimental Results Demonstrate A 45% Reduction In Debugging Time And A 60% Improvement In Code Comprehension Among Development Teams. The Platform Supports Multiple Programming Languages And Integrates Seamlessly With Existing Development Environments. Read more Read less

SAFEHEAVEN – WOMEN SAFETY APPLICATION USING REAL-TIME INTELLIGENT ALERT SYSTEM

Women’s Safety Is A Critical Issue In Modern Society, Especially In Urban And Semi-urban Environments Where Immediate Assistance Is Often Delayed. This Paper Presents SafeHeaven, A Mobile-based Women Safety Application Designed To Provide Real-time E... Women’s Safety Is A Critical Issue In Modern Society, Especially In Urban And Semi-urban Environments Where Immediate Assistance Is Often Delayed. This Paper Presents SafeHeaven, A Mobile-based Women Safety Application Designed To Provide Real-time Emergency Response Using Advanced Technologies Such As GPS Tracking, Firebase Cloud Services, And Mobile Sensors. The System Enables Users To Trigger SOS Alerts Through Multiple Methods Including One-tap Activation, Shake Detection, And Voice Commands. Once Activated, The System Shares Live Location Data With Nearby Officers, Police Stations, And Emergency Contacts. Additional Features Such As Geo-fencing, Panic Alarms, And Fake Call Simulation Enhance Preventive Safety. The Backend Is Implemented Using Firebase Authentication, Firestore Database, And Cloud Messaging, Ensuring Scalability And Real-time Communication. The Proposed System Improves Response Time, Enhances Coordination, And Provides A Reliable Safety Solution For Women. Read more Read less

Portable SOC Log Analyzer For Isolated Networks (Offline SIEM Lite)

The Growing Complexity Of Cyber Threats Has Increased The Importance Of Continuous Monitoring And Analysis Of System Logs. Analyzing System Logs Reveals Patterns Of Activity That Help Uncover Unusual Behaviour And Possible Security Issues. However, M... The Growing Complexity Of Cyber Threats Has Increased The Importance Of Continuous Monitoring And Analysis Of System Logs. Analyzing System Logs Reveals Patterns Of Activity That Help Uncover Unusual Behaviour And Possible Security Issues. However, Most Modern Security Information And Event Management (SIEM) Solutions Rely On Cloud Infrastructure And Require Constant Internet Connectivity, Which Limits Their Usability In Isolated Or Restricted Environments. This Paper Introduces A Portable SOC Log Analyzer, A Lightweight And Offline-capable System Designed To Perform Log Analysis Without External Dependencies. The System Collects Logs From Multiple Sources, Processes Them Into Structured Formats, And Applies Rule-based Detection Techniques To Identify Suspicious Activities Such As Repeated Login Failures, Unauthorized Access Attempts, And Abnormal Behaviour Patterns. The Proposed Solution Includes Features Such As Alert Generation, Log Filtering, And Graphical Visualization Through A User-friendly Interface. Because It Operates Fully Offline, The Tool Preserves Privacy And Remains Dependable, Which Makes It Well-suited For Isolated Or Air‑gapped Systems. Read more Read less

SMART MULTI-RESTAURANT FOOD ORDERING AND DELIVERY SYSTEM

The Smart Multi-Restaurant Food Ordering And Delivery System Is An Advanced Food Delivery Platform Built To Overcome The Limitations Of Traditional Single-restaurant Ordering Applications. The System Allows Users To Order From Multiple Restaurants In... The Smart Multi-Restaurant Food Ordering And Delivery System Is An Advanced Food Delivery Platform Built To Overcome The Limitations Of Traditional Single-restaurant Ordering Applications. The System Allows Users To Order From Multiple Restaurants In A Single Transaction, Eliminating Multiple Payments, Separate Order Tracking, And Redundant Delivery Charges. The System Is Developed Using React.js As The Frontend Framework Deployed On Vercel, Python Flask As The Backend REST API, And PostgreSQL As The Relational Database. Payment Processing Is Handled Through Razorpay, And Location Services Including Restaurant Discovery, Delivery Tracking, And Route Optimization Are Powered By The Mapbox API. The Platform Integrates Artificial Intelligence And Machine Learning Technologies For Food Pairing Recommendations Using Collaborative Filtering, And A Convolutional Neural Network(CNN) Based Hygiene Monitoring System Achieving Approximately 91% Classification Accuracy. A Multi-stop Route Optimization Module Ensures Efficient Delivery Path Planning For Orders Spanning Multiple Restaurants. The System Adopts A Three-tier Architecture Comprising A React.js Presentation Layer, A Python Flask Application Layer, And A PostgreSQL Data Layer. Key Features Include A Smart Multi-Restaurant Cart With Single-checkout, AI-powered Recommendations, Realtime Mapbox Order Tracking, And JWT-based Role-based Access Control. The Platform Supports 10,000 Or More Concurrent Users With A 99.5% Uptime Target. Read more Read less

Digital Asset Discovery For Organization

The Rapid Expansion Of Cloud Computing, APIs, And Third-party Services Has Significantly Increased The Digital Attack Surface Of Modern Organizations. Many Cyber Incidents Occur Due To Unmanaged Or Forgotten Assets Such As Test Servers, Misconfigured... The Rapid Expansion Of Cloud Computing, APIs, And Third-party Services Has Significantly Increased The Digital Attack Surface Of Modern Organizations. Many Cyber Incidents Occur Due To Unmanaged Or Forgotten Assets Such As Test Servers, Misconfigured Cloud Storage, And Undocumented APIs. Existing Security Tools Often Lack Continuous External Visibility And Centralized Asset Mapping. This Paper Presents A Digital Asset Discovery Framework That Automatically Identifies And Monitors Organizational Digital Assets From An Attacker’s Perspective. The System Performs Automated Enumeration Of Domains, Subdomains, IP Addresses, Cloud Resources, APIs, And Certificates. Discovered Assets Are Enriched With Exposure And Hosting Information, Followed By Risk Classification And Prioritization. Continuous Monitoring Mechanisms Detect Newly Exposed Assets In Real Time. The Proposed Approach Improves Asset Visibility, Reduces Security Blind Spots, And Supports Proactive Security Management Through Dashboards And Reports. Read more Read less

DESIGN OF VLSI-BASED HARDWARE DETECTION SYSTEM AGAINST POWER SIDE-CHANNEL ATTACKS

The Rapid Growth Of System-on-Chip (SoC) Technologies Has Increased Their Vulnerability To Advanced Physical And Power Side-channel Attacks, Which Can Compromise The Security Of Integrated Circuits. Conventional Countermeasures Often Involve Complex... The Rapid Growth Of System-on-Chip (SoC) Technologies Has Increased Their Vulnerability To Advanced Physical And Power Side-channel Attacks, Which Can Compromise The Security Of Integrated Circuits. Conventional Countermeasures Often Involve Complex Fabrication Processes And High Area Overhead, Making Them Difficult To Implement And Verify. This Paper Presents A VLSI-based Hardware Detection System Designed To Identify And Mitigate Power Side-channel And Physical Attacks In Real Time. The Proposed Approach Integrates Clock-based And Voltage-based Sensing Mechanisms Along With Dynamic Pattern Matching To Detect Anomalies Caused By Attacks Such As Voltage Manipulation, Replay Attacks, And Unauthorized Probing. The System Is Implemented Using FPGA-based Architecture And Validated Through Simulation Using ModelSim. Experimental Results Demonstrate That The Proposed Detection Framework Achieves High Accuracy With Low Overhead, Ensuring Minimal Impact On System Performance. The Design Offers A Scalable And Efficient Solution That Can Be Seamlessly Integrated Into Existing SoC Architectures, Enhancing Hardware Security And Reliability In Modern Embedded Systems. Read more Read less

HELMET AND NUMBER PLATE DETECTION USING DEEP LEARNING

The Project Titled "Helmet And Number Plate Detection Using Deep Learning" Employs Advanced Computer Vision And Deep Learning Techniques To Enhance Road Safety And Automate Traffic Law Enforcement. The System Is Developed Using Python And Utilizes Th... The Project Titled "Helmet And Number Plate Detection Using Deep Learning" Employs Advanced Computer Vision And Deep Learning Techniques To Enhance Road Safety And Automate Traffic Law Enforcement. The System Is Developed Using Python And Utilizes The YOLOv8 (YouOnly Look Once, Version 8) Architecture — A State-of-the-art Object Detection Model — For Real-time Identification Of Helmets And Vehicle Number Plates. The Web-based Interface Is Created Using HTML, CSS, And JavaScript, Supported By The Flask Framework, Ensuring Responsive And User-friendly Interaction. The YOLOv8 Model Is Trained On A Comprehensive Dataset Containing Various Road Scenes Under Diverse Lighting, Weather, And Camera Conditions. The Model Achieved A Training Accuracy Of88% And A Validation Accuracy Of 79%, Indicating Strong Performance In Detecting Both Helmets And Number Plates. The System Operates In Three Modes: Image Mode, Video Mode, And Web Camera Mode, Providing Flexibility For Static And Real-time Detection. It Automates The Identification Process, Helping Traffic Authorities Monitor Compliance And Improve Public Safety. This Project Thus Demonstrates The Effective Integration Of Deep Learning With Computer Vision For Intelligent Transportation Monitoring And Enforcement. Read more Read less

A Hybrid Deep Learning Model For Detecting Credit Card Fraud Using CNN–BiLSTM With An Attention Mechanism And Focal Loss Optimization

Detecting Credit Card Fraud Is A Complex Task Due To The Significant Imbalance In Data And The Constantly Changing Nature Of Fraudulent Activities. This Research Introduces A Hybrid Deep Learning Model That Combines CNN, BiLSTM, And An Attention Mech... Detecting Credit Card Fraud Is A Complex Task Due To The Significant Imbalance In Data And The Constantly Changing Nature Of Fraudulent Activities. This Research Introduces A Hybrid Deep Learning Model That Combines CNN, BiLSTM, And An Attention Mechanism To Effectively Identify Spatial And Temporal Patterns In Transactions. To Maintain Regional Specificity, Separate Models Were Developed For Datasets From India And Europe. To Tackle Class Imbalance Without Introducing Synthetic Bias, Focal Loss With Adaptive Class Weighting Was Employed. The Experiments Revealed That The Model For The European Dataset Achieved An Accuracy Of 99.32% And An F1-score Of 97.29%, While The Model For The Indian Dataset Reached An Accuracy Of 98.67% And An F1-score Of 95.80%. The Use Of Attention Mechanisms Enhanced The Relevance Of Features And Overall Performance, And SHAP-based Explainability Improved The Interpretability Of The Model. The System Was Deployed Using Stream Lit For Real-time Fraud Detection, Offering A Scalable And Precise Solution Tailored To Region-specific Financial Fraud Detection. Read more Read less

AUTOMATIC PLATFORM BRIDGE CONNECTOR IN RAILWAYS USING EMBEDDED SYSTEM

The Project “Automatic Platform Bridge Connector In Railways Using Embedded System” Is Designed To Improve Passenger Safety And Operational Efficiency At Railway Stations. The System Utilizes An Arduino Microcontroller To Control A DC Motor That Auto... The Project “Automatic Platform Bridge Connector In Railways Using Embedded System” Is Designed To Improve Passenger Safety And Operational Efficiency At Railway Stations. The System Utilizes An Arduino Microcontroller To Control A DC Motor That Automatically Connects Two Platforms Through A Movable Bridge. Infrared (IR) Sensors Detect The Arrival And Departure Of Trains, Triggering Automatic Bridge Operation To Ensure Safe Timing. An LED Indicator Alerts Passengers About Approaching Trains, While A Relay Module Ensures Precise Motor Control For Bridge Movement. A 16x2 LCD Display Provides Real-time Updates On Train Detection And Bridge Status. By Eliminating Manual Operation, This System Reduces The Risk Of Accidents, Enhances Passenger Convenience, And Ensures Safe And Intelligent Platform Connectivity In Railway Environments. The System’s Design Emphasizes Automation, Reliability, And Safety Integration. When An Incoming Train Is Detected By The IR Sensors, The Arduino Immediately Deactivates The Bridge By Controlling The DC Motor Through The Relay Module, Preventing Passengers From Crossing. Once The Train Safely Departs And The Sensors Confirm A Clear Track, The Arduino Reactivates The Motor To Reconnect The Bridge Automatically, Allowing Safe Passage Between Platforms. The LCD Display Continuously Updates The Bridge And Train Status, While The LED Indicator Visually Warns Passengers Of Train Movements. This Automated Mechanism Not Only Minimizes Human Error But Also Ensures Efficient Station Management, Making It A Cost-effective And Intelligent Solution For Modern Railway Infrastructure. Read more Read less

BLOOD GROUP DETECTION USING FINGERPRINT

Fingerprint-based Blood Group Detection Is An Innovative And Emerging Technique That Integrates Deep Learning And Image Processing To Predict An Individual's Blood Type From Fingerprint Patterns. This Method Leverages The Correlation Between Fingerpr... Fingerprint-based Blood Group Detection Is An Innovative And Emerging Technique That Integrates Deep Learning And Image Processing To Predict An Individual's Blood Type From Fingerprint Patterns. This Method Leverages The Correlation Between Fingerprint Ridge Characteristics And Blood Group–related Antigens Secreted Through Sweat Glands. After Obtaining The Fingerprint Image, Preprocessing Techniques Are Applied To Enhance Its Quality, Followed By Feature Extraction Using Convolutional Neural Networks (CNNs) And Other Machine Learning Models. The Trained Model Then Classifies The Fingerprint Into Specific Blood Groups Such As A+, B-, O+, Etc. The Proposed Method Aims To Provide A Non-invasive, Fast, And Portable Alternative To Traditional Serological Blood Tests. This Approach Has Potential Applications In Emergency Healthcare, Forensic Science, And Medical Diagnostics. Read more Read less

COLLEGE MANAGEMENT SYSTEM

The College Management System Developed Using The MERN Stack (MongoDB, Express.js, React.js, And Node.js) Is A Comprehensive Web-based Application Designed To Automate And Streamline Academic And Administrative Activities In Higher Education Institut... The College Management System Developed Using The MERN Stack (MongoDB, Express.js, React.js, And Node.js) Is A Comprehensive Web-based Application Designed To Automate And Streamline Academic And Administrative Activities In Higher Education Institutions. The System Provides A Centralized Platform For Managing Student Information, Staff Operations, Attendance, Marks, Class Schedules, And Study Materials With Secure Role-based Access For Students, Staff, And Admin. It Replaces Traditional Manual And Semi-digital Processes With A Modern, Scalable, And User-friendly Solution That Ensures Data Accuracy, Transparency, And Real-time Accessibility. The Application Supports JWT-based Authentication For Enhanced Security, Structured Data Storage Using MongoDB, And Responsive Interfaces Using React And Bootstrap For Both Desktop And Mobile Views. Key Features Include Semester-wise Marks Management With CGPA Calculation, Attendance Tracking With Percentage Computation, Upload And Distribution Of Study Materials, Class Schedule Management, And Generation Of Official Documents Such As Transcripts And Attendance Reports In PDF Format. By Integrating Automation, Secure Data Handling, And Modular Design, The Proposed System Significantly Reduces Administrative Workload, Minimizes Human Errors, And Improves Communication Between Students And Faculty, Making It A Reliable And Efficient Digital Solution For Modern College Management. Read more Read less

ASSESMENT OF GROUNDWATER CONTAMINATION NEAR LANFILLS

Groundwater Is A Critical Source Of Drinking Water Worldwide, But It Is Increasingly Threatened By Contamination From Landfill Sites. Landfills Generate Leachate—a Toxic Liquid Formed By The Percolation Of Water Through Waste—which Can Infiltrate Aqu... Groundwater Is A Critical Source Of Drinking Water Worldwide, But It Is Increasingly Threatened By Contamination From Landfill Sites. Landfills Generate Leachate—a Toxic Liquid Formed By The Percolation Of Water Through Waste—which Can Infiltrate Aquifers And Degrade Groundwater Quality. This Paper Reviews The Sources, Mechanisms, Assessment Methods, And Impacts Of Groundwater Contamination Near Landfills. It Also Discusses Monitoring Techniques And Mitigation Strategies. Studies Show That Contamination Is Most Severe Within 200–1000 M Of Landfill Sites And Includes Heavy Metals, Organic Pollutants, And Microbial Contaminants. Effective Landfill Management And Monitoring Systems Are Essential To Protect Groundwater Resources. Read more Read less

VISION-DRIVEN FRAMEWORK FOR STUDENT IDENTITY VERIFICATION USING FACE AS HALL TICKET IN MODERN EXAMINATION SYSTEMS

Student Authentication During Examinations Has Emerged As A Critical Challenge In Modern Academic Environments Due To The Rising Incidence Of Impersonation, Identity Fraud, And Unethical Examination Practices. Conventional Verification Methods Such A... Student Authentication During Examinations Has Emerged As A Critical Challenge In Modern Academic Environments Due To The Rising Incidence Of Impersonation, Identity Fraud, And Unethical Examination Practices. Conventional Verification Methods Such As Hall Tickets, Identity Cards, And Manual Invigilation Are Highly Vulnerable On Document-based Verification And Basic Biometric Techniques Suffer From Poor Scalability, Limited Real-time Effectiveness, And An Inability To Accurately Confirm The Physical Presence Of Candidates, Particularly Under Uncontrolled Examination Conditions. To Address The Limitations Of Traditional Examination Authentication Methods, This Project Proposes A Deep Learning–based Automated Student Authentication System In Which The Student’s Face Acts As A Digital Hall Ticket. The Framework Employs Multi-Task Cascaded Convolutional Neural Networks (MTCNN) For Accurate Facial Detection And Alignment, Enabling Reliable Face Localization Under Varying Pose And Lighting Conditions Commonly Found In Examination Halls. To Ensure That The Candidate Is Physically Present And To Prevent Spoofing Attacks Using Photographs Or Recorded Videos, A CNN-based Liveness Verification Module Is Integrated. After Successful Liveness Validation, FaceNet Is Use Live Facial Inputs With Enrolled Student Data. Upon Authentication, The System Automatically Displays The Student’s Hall Name, Seat Number, Examination Details, And Assigned Invigilator. During The Examination, The Invigilator Captures The Student’s Face, And The System Verifies The Hall Ticket In Real Time. If Impersonation Is Detected, An Instant SMS Alert Is Sent To The Controller Of Examinations And The Respective Head Of Department. Overall, The System Strengthens Examination Security, Prevents Impersonation, And Automates Hall Ticket Verification. Read more Read less

LIFESCAN: REAL-TIME SURVIVOR DETECTION USING NON-CONTACT VITAL SIGN MONITORING AND DEEP LEARNING

Lifescan Offers A Non-invasive, Efficient, And Reliable Approach To Locating Survivors, Significantly Reducing Rescue Time And Improving The Chances Of Saving Lives. This Technology Has Potential Applications In Disaster Management, Military Operatio... Lifescan Offers A Non-invasive, Efficient, And Reliable Approach To Locating Survivors, Significantly Reducing Rescue Time And Improving The Chances Of Saving Lives. This Technology Has Potential Applications In Disaster Management, Military Operations, And Remote Healthcare MonitoringLifescan: Real-Time Survivor Detection Using Non-Contact Vital Sign Monitoring And Deep Learning Is An Advanced System Designed To Enhance Search And Rescue Operations In Disaster Scenarios. The Proposed System Utilizes Non-contact Sensing Technologies Such As Optical Cameras, Thermal Imaging, And Radar Sensors To Detect Human Vital Signs, Including Heart Rate And Respiration, Without Requiring Physical Contact. These Physiological Signals Are Often Difficult To Capture In Challenging Environments Such As Collapsed Structures Or Low-visibility Conditions.To Address This, The System Integrates Deep Learning Algorithms For Accurate Detection And Classificationof Human Presence Based On Extracted Vital Signals. The Collected Data Is Processed Through Signal Filtering And Feature Extraction Techniques, And Then Analyzed Using Trained Neural Network Models To Determine The Likelihood Of Survival. By Combining Sensor Data With Intelligent Analysis, The System Can Identify Survivors In Real Time, Even If They Are Unconscious Or Immobile.Lifescan Offers A Non-invasive, Efficient, And Reliable Approach To Locating Survivors, Significantly Reducing Rescue Time And Improving The Chances Of Saving Lives. This Technology Has Potential Applications In Disaster Management, Military Operations, And Remote Healthcare Monitoringthe Proposed System Emphasizes Robustness And Adaptability In Dynamic And Noisy Environments Commonly Encountered During Disaster Situations. Advanced Preprocessing Techniques Are Employed To Minimize Interference Caused By Dust, Debris, And Environmental Disturbances, Ensuring Reliable Signal Acquisition. The Deep Learning Models Are Trained On Diverse Datasets To Improve Generalization And Accuracy Across Different Scenarios, Including Varying Lighting Conditions And Partial Occlusions. The Conductected Source Can Developed. Read more Read less

Multimodal Fake News Detection Using DistilBERT And Xception Convolutional Neural Network

The Rapid Proliferation Of Misinformation Across Digital Platforms Poses A Growing Threat To Public Discourse, Democratic Processes, And Societal Trust. Traditional Fake News Detection Approaches Rely Exclusively On Textual Analysis, Failing To Explo... The Rapid Proliferation Of Misinformation Across Digital Platforms Poses A Growing Threat To Public Discourse, Democratic Processes, And Societal Trust. Traditional Fake News Detection Approaches Rely Exclusively On Textual Analysis, Failing To Exploit The Deceptive Potential Of Associated Images. This Paper Proposes A Multimodal Deep Learning Framework That Combines DistilBERT For Efficient Semantic Text Analysis And The Xception Convolutional Neural Network For Visual Feature Extraction. DistilBERT, Derived From BERT Through Knowledge Distillation, Retains 97% Of BERT’s Language Understanding While Reducing Model Size By 40% And Improving Inference Speed. Xception Leverages Depthwise Separable Convolutions To Extract Discriminative Visual Patterns From News Images. Features From Both Modalities Are Concatenated Through A Fully Connected Fusion Layer And Classified Using Sigmoid Activation. The Proposed System Is Evaluated On The FakeNewsNet And GossipCop Datasets, Achieving 93.47% Accuracy With An F1-score Of 0.93, Outperforming Single-modality And Several Prior Multimodal Baselines. A Flask-based REST API Enables Real-time Deployment With Confidence Score Output. Read more Read less

Driver Drowsiness Detection Using Machine Learning

Driver Drowsiness Is One Of The Leading Causes Of Road Accidents Worldwide, Posing A Significant Threat To Human Life And Road Safety. This Paper Presents A Real-time Driver Drowsiness Detection System Employing Convolutional Neural Networks (CNNs) T... Driver Drowsiness Is One Of The Leading Causes Of Road Accidents Worldwide, Posing A Significant Threat To Human Life And Road Safety. This Paper Presents A Real-time Driver Drowsiness Detection System Employing Convolutional Neural Networks (CNNs) To Analyze Facial Cues Captured Via An In-vehicle Camera. The System Monitors Critical Fatigue Indicators Including Eye Closure Rate, Blink Frequency, Mouth State (yawning), And Head Orientation. A Modular Pipeline — Encompassing Image Acquisition, Face Detection Using Haar Cascades/SSD, Facial Landmark Extraction With OpenCV/Dlib, And CNN-based Drowsiness Classification — Enables Robust Real-time Inference. Upon Detecting Drowsiness, The System Triggers Multi-modal Alerts To Prompt Corrective Driver Action. Experimental Results Demonstrate High Detection Accuracy Across Varied Lighting And Environmental Conditions, Making The Proposed System A Viable Enhancement For Modern Vehicle Safety Systems. Read more Read less

AI-Powered Twitter Content Moderation Using SBERT, CNN And Bi-LSTM With Severity-Based Alert System

Social Media Platforms Such As Twitter (now X) Generate Large Volumes Of Real-time Content, Including Spam, Malicious Links, And Harmful Messages. The Rapid Spread Of Such Content Poses Significant Challenges For Manual Moderation Due To High Data Ve... Social Media Platforms Such As Twitter (now X) Generate Large Volumes Of Real-time Content, Including Spam, Malicious Links, And Harmful Messages. The Rapid Spread Of Such Content Poses Significant Challenges For Manual Moderation Due To High Data Velocity And Evolving Patterns Of Misuse. This Paper Proposes An AI-powered Twitter Content Moderation System That Integrates Semantic Analysis, Deep Learning, And Rule-based Validation For Effective Detection Of Harmful Content. The Proposed System Consists Of A Multi-stage Pipeline. Ini-tially, A URL Threat Detection Module Is Employed To Identify Suspicious Links Such As Malicious Domains And Shortened URLs. Cleaned Tweet Content Is Then Processed Using Sentence-BERT (SBERT) To Generate Semantic Embeddings. These Embeddings Are Passed Through A Hybrid Deep Learning Model Combining Convolutional Neural Networks (CNN) For Local Feature Extrac-tion And Bidirectional Long Short-Term Memory (Bi-LSTM) Networks For Capturing Contextual Dependencies. The Model Classifies Tweets Into Categories Such As Malicious, Spam, And Non-spam. To Enhance Reliability, Additional Decision Layers Including Confidence Threshold Checks And Rule-based Overrides Are In-corporated To Handle Uncertain Predictions And Known Spam Patterns. Furthermore, A Severity-based Alert Mechanism Is Im-plemented To Trigger Real-time Notifications For High-risk Content. Experimental Evaluation Demonstrates That The Proposed Hybrid Model Improves Classification Accuracy, Robustness, And Real-time Applicability Compared To Traditional And Standalone Approaches. Read more Read less

Smart Coastal Sprayer For Small Scale Farmers Using Wind-Assisted Battery Charging

Traditional Pesticide Spraying In Coastal Regions Suffers 40-60% Wastage Due To Unpredictable Wind Patterns. This Paper Introduces An ESP32-based Autonomous Sprayer Integrating Wind Turbine Battery Charging, Real-time Anemometer Monitoring, And MPU60... Traditional Pesticide Spraying In Coastal Regions Suffers 40-60% Wastage Due To Unpredictable Wind Patterns. This Paper Introduces An ESP32-based Autonomous Sprayer Integrating Wind Turbine Battery Charging, Real-time Anemometer Monitoring, And MPU6050 Stabilization. The Wind-aware Adaptive Algorithm Dynamically Adjusts Nozzle Angle And Spray Rate Based On Wind Velocity/direction, Achieving 42% Chemical Savings And 91% Coverage Uniformity. The System Enables Small-scale Farmers (<5 Acres) To Access Precision Agriculture Technology While Ensuring Sustainability Through Renewable Wind Energy Harvesting. Read more Read less

Face Identification In Movie Footage

Face Identification And Object Detection In Movie Footage Present Significant Challenges Owing To The Dynamic Nature Of Video Content, Complex Scene Transitions, Rapid Movements, And Occlusions. Existing Systems That Apply Image-based Algorithms Fram... Face Identification And Object Detection In Movie Footage Present Significant Challenges Owing To The Dynamic Nature Of Video Content, Complex Scene Transitions, Rapid Movements, And Occlusions. Existing Systems That Apply Image-based Algorithms Frame-by-frame Are Computationally Intensive, Lack Temporal Consistency, And Fail To Exploit Contextual Information Across Consecutive Frames. This Paper Proposes A Two-stage Deep Learning Pipeline For Face Identification And Object Detection In Movie Footage. The First Stage Employs A Region Proposal Network To Identify Candidate Regions With High Recall, While The Second Stage Applies A Deep Learning-based Classifier To Assign Categorical Labels And Refine Bounding Box Localization. The System Design Is Presented Through Data Flow And Database Design Diagrams, Demonstrating A Structured Pipeline From Image Capture Through Face Detection, Alignment, Feature Extraction, To Final Feature Matching And Application Control. Read more Read less

E-Voting Using Face Recognition

This Paper Presents The Design And Implementation Of An E-Voting System Using Face Recognition Technology. The Proposed System Addresses Key Challenges In Traditional Voting Methods, Including Voter Impersonation, Manual Errors, And Slow Vote Countin... This Paper Presents The Design And Implementation Of An E-Voting System Using Face Recognition Technology. The Proposed System Addresses Key Challenges In Traditional Voting Methods, Including Voter Impersonation, Manual Errors, And Slow Vote Counting. By Integrating Facial Biometrics, Encrypted Data Storage, And An Automated Result Generation Mechanism, The System Ensures A Secure, Transparent, And Efficient Electoral Process. The System Is Built With Five Core Modules — Admin, Field Officer, Nominee, Voter, And Result — And Employs Image Processing Techniques Such As Face Detection, Alignment, Feature Extraction, And Feature Matching To Authenticate Voters In Real Time. Experimental Results Demonstrate That The Proposed System Is Both Technically Feasible And Socially Acceptable, Offering A Reliable Alternative To Conventional Voting Systems. Read more Read less

A PROXY RE-ENCRYPTION APPROACH TO SECURE DATA SHARING IN THE INTERNET OF THINGS BASED ON BLOCKCHAIN

The Rapid Evolution Of The Internet Of Things (IoT) Has Enabled Massive Data Generation And Sharing Across Cloud Platforms. However, Data Security Remains A Critical Challenge, As Unauthorized Access Or Misuse Can Lead To Severe Consequences. To Addr... The Rapid Evolution Of The Internet Of Things (IoT) Has Enabled Massive Data Generation And Sharing Across Cloud Platforms. However, Data Security Remains A Critical Challenge, As Unauthorized Access Or Misuse Can Lead To Severe Consequences. To Address This, We Propose A Proxy Re-Encryption (PRE) Based Approach For Secure Data Sharing In IoT Environments, Integrated With Blockchain For Decentralization And Trust. Data Owners Can Encrypt Their Data Using Identity-based Encryption And Outsource It To The Cloud. A Proxy Server, Typically An Edge Device, Handles Intensive Computations To Re-encrypt The Data For Authorized Users, Enabling Secure Access Without Revealing The Original Content. Blockchain Is Integrated To Record Data Access Requests, Approvals, And Transactions In A Tamper-proof And Decentralized Manner, Ensuring Transparency And Trust. By Splitting Data Into Multiple Blocks And Enforcing Trusted Authority Validation For Users And Owners, The System Enhances Confidentiality, Integrity, And Availability. Overall, This Approach Provides A Scalable, Efficient, And Secure Solution For IoT Data Sharing With Minimal Delay And Reduced Computational Overhead On IoT Devices. Read more Read less

A Comparative Design And Analysis Of Pre-Engineered Building (PEB) And Conventional Building (CEB)

This Paper Mainly Focuses On The PEB Concept And CSB Concept. The Pre- Engineered Building (PEB) Concept Is A New Conception Of Single-story Industrial Building Construction. This Methodology Is Versatile Due To Its Lightweight And Economical Constru... This Paper Mainly Focuses On The PEB Concept And CSB Concept. The Pre- Engineered Building (PEB) Concept Is A New Conception Of Single-story Industrial Building Construction. This Methodology Is Versatile Due To Its Lightweight And Economical Construction. This Concept Has Many Advantages Over The Conventional Steel Building (CSB) Concept Of Buildings With Roof Trusses. In This Work, An Industrial Building, Of Length 21m And Width Is83m. The Slope Of Roof Trusses Is Taken As 5.71degree. Eave Height Is 6m. These Structures Havebeenanalyzed Anddesigned ByusingSTAAD Pro V8i To Compare The PEB And Conventional Steel Truss. PEB Design Is Based On The American Code AISC 360:10 And CSB Design Is Based On The Indian Code IS800:2007. Loads Considered In The Analysis Are Dead Load, Live Load, And Wind Load Along With The Various Combinations As Specified In IS800:2007and AISC.Deadload Istakenbasedon IS: 875 (Part 1)-1987. Live Load Is Taken Based On IS: 875(Part-2)-1987. Wind Load Is Taken Based On IS: 875 (Part 3)-2015. The Structure Is Located At Savner In The Nagpur District. Read more Read less

WATER QUALITY AND WATER LEAKAGE MONITORING SYSTEM USING IOT

This Paper Presents The Design And Implementation Of An IoT-based Water Quality And Water Leakage Monitoring System. The System Utilizes An Arduino Uno Microcontroller Integrated With Turbidity, Flow, And Level Sensors To Continuously Monitor Water Q... This Paper Presents The Design And Implementation Of An IoT-based Water Quality And Water Leakage Monitoring System. The System Utilizes An Arduino Uno Microcontroller Integrated With Turbidity, Flow, And Level Sensors To Continuously Monitor Water Quality Parameters And Detect Leakages In Real Time. When Contamination Or Leakage Is Detected, The System Triggers An Alarm And Activates A Solenoid Valve Through A Relay To Shut Off The Water Supply Automatically. Sensor Data Is Transmitted To The Cloud Via IoT, Enabling Remote Monitoring Through An Android Application. The Proposed System Offers A Cost-effective, Automated Solution For Ensuring Safe Water Distribution And Minimizing Water Wastage. Read more Read less

DPDP Compliance Tool

This Project Presents The Design And Development Of A DPDP Compliance Tool Aimed At Helping Organizations Manage Personal Data In Accordance With The Digital Personal Data Protection Act, 2023. With Increasing Concerns Over Data Privacy, Organization... This Project Presents The Design And Development Of A DPDP Compliance Tool Aimed At Helping Organizations Manage Personal Data In Accordance With The Digital Personal Data Protection Act, 2023. With Increasing Concerns Over Data Privacy, Organizations Must Ensure That User Data Is Collected, Processed, And Stored In A Transparent And Lawful Manner. The System Focuses On Two Key Modules: Cookie Consent Management And User Consent Management, Enabling Websites To Obtain, Record, And Manage User Consent Effectively. The Cookie Module Allows Users To Accept, Reject, Or Customize Their Preferences Regarding Tracking Technologies, While The User Consent Module Ensures Explicit Consent Is Obtained For Data Processing And Provides Options To Review, Update, Or Withdraw Consent At Any Time. Additionally, The Tool Supports Audit Trails, Role-based Access Control, And Compliance Reporting To Ensure Accountability And Traceability. Developed Using Modern Technologies And Secure Practices, This Solution Enhances Data Protection, Builds User Trust, And Helps Organizations Meet Regulatory Compliance Requirements Efficiently. Read more Read less

Crop Prediction And Soil Nutrient Monitoring System

Precision Agriculture Plays A Crucial Role In Improving Crop Productivity While Promoting Efficient Use Of Resources. This Project Tackles Issues Such As Inconsistent Crop Yield And Improper Fertiliser Usage By Designing An Integrated System For Crop... Precision Agriculture Plays A Crucial Role In Improving Crop Productivity While Promoting Efficient Use Of Resources. This Project Tackles Issues Such As Inconsistent Crop Yield And Improper Fertiliser Usage By Designing An Integrated System For Crop Prediction And Soil Nutrient Monitoring. The Proposed System Is Based On Internet Of Things (IoT) Technology, Where Wireless Sensors Are Installed In Agricultural Fields To Continuously Monitor Important Soil Nutrients, Including Nitrogen (N), Phosphorus (P), And Potassium (K). The Collected Data, Along With Environmental Parameters, Is Sent To A Cloud Platform For Analysis. Machine Learning Techniques, Such As Random Forest Or Support Vector Machine (SVM), Are Then Used To Recommend Suitable Crops And Forecast Expected Yields By Analysing Both Real-time Inputs And Past Climatic Trends. Additionally, The System Generates Location-specific Suggestions For Fertiliser Application And Irrigation Management, Replacing Conventional Uniform Practices With A More Precise And Data-driven Approach. The Main Goal Is To Support Farmers In Making Informed Decisions, Ultimately Increasing Productivity, Lowering Operational Costs, And Reducing Environmental Damage Caused By Excessive Use Of Fertiliser. Overall, This System Aims To Enhance Agricultural Performance While Ensuring Sustainable Management Of Natural Resources. Read more Read less

DEEP LEARNING-DRIVEN PREDICTIVE MAINTENANCE FOR ARTIFICIAL YARN MACHINE WITH REAL-TIME IOT DEPLOYMENT

Predictive Maintenance Has Emerged As A Critical Requirement In Modern Textile Manufacturing To Minimize Machine Downtime And Improve Operational Efficiency. This Research Presents A Deep Learning–based Predictive Maintenance System For Artificial Ya... Predictive Maintenance Has Emerged As A Critical Requirement In Modern Textile Manufacturing To Minimize Machine Downtime And Improve Operational Efficiency. This Research Presents A Deep Learning–based Predictive Maintenance System For Artificial Yarn Machines Integrated With Real-time IoT Deployment. The System Collects Sensor Data Such As Temperature, Vibration, And Operational Load From Yarn Machines Continuously. Advanced Data Preprocessing Techniques Are Applied To Clean And Structure The Incoming Data Stream. A Deep Learning Model Is Trained To Identify Hidden Patterns Associated With Machine Failures. The Proposed System Predicts Potential Faults Before They Occur, Enabling Proactive Maintenance. Real-time Monitoring Is Achieved Through IoT Devices Connected To A Centralized Analytics Platform. The System Reduces Unexpected Breakdowns And Enhances Machine Lifespan. It Also Improves Production Quality And Consistency In Yarn Manufacturing. The Model Is Evaluated Using Multiple Performance Metrics To Ensure Reliability. Experimental Results Demonstrate Improved Prediction Accuracy Compared To Traditional Methods. The Solution Is Scalable And Suitable For Industrial Deployment. This Work Contributes To Intelligent Manufacturing By Integrating AI And IoT Technologies Effectively. Read more Read less

Privacy-Preserving Multiple Payment Fraud Detection Using LSTM-based Federated Learning.

This Paper Presents A Privacy-preserving Framework For Multiple Payment Fraud Detection Using Long Short-Term Memory (LSTM) And Federated Learning. The Model Captures Temporal Patterns In Transaction Data To Identify Fraudulent Activities Effectively... This Paper Presents A Privacy-preserving Framework For Multiple Payment Fraud Detection Using Long Short-Term Memory (LSTM) And Federated Learning. The Model Captures Temporal Patterns In Transaction Data To Identify Fraudulent Activities Effectively. Federated Learning Enables Decentralized Training Across Multiple Clients Without Sharing Sensitive Data, Ensuring Privacy And Security. Experimental Results Show That The Proposed Approach Achieves Over 95% Accuracy With Improved Precision And Recall Compared To Traditional Methods. Additionally, It Reduces Data Leakage Risks While Maintaining Scalability And Efficiency. The Proposed System Is Suitable For Real-time Fraud Detection In Distributed Financial Environments. Read more Read less

TRANSGUARD-RT: A REAL-TIME TRANSFORMER-DRIVEN INTRUSION DETECTION SYSTEM USING LIVE NETWORK TRAFFIC ANALYTICS

The Rapid Expansion Of Internet-based Communication And Interconnected Digital Infrastructures Has Significantly Increased Exposure To Advanced Cyber Threats, Including Malware Infiltration, Phishing Attempts, Denial-of-service Attacks, And Unauthori... The Rapid Expansion Of Internet-based Communication And Interconnected Digital Infrastructures Has Significantly Increased Exposure To Advanced Cyber Threats, Including Malware Infiltration, Phishing Attempts, Denial-of-service Attacks, And Unauthorized Access. Traditional Security Mechanisms Such As Signature-based Intrusion Detection And Rule-driven Firewalls Are Often Insufficient To Identify Evolving And Previously Unseen Attack Patterns, Particularly In High-speed And Complex Network Environments. To Address These Limitations, This Research Presents A Real-time Intrusion Detection System That Leverages Live Network Traffic Captured Through NCAP- Based Monitoring And Applies A Transformer-based Deep Learning Architecture For Intelligent Threat Detection. The Proposed Approach Focuses On Analyzing Sequential Dependencies And Contextual Relationships With In Network Traffic Data Using Self-attention Mechanisms, Enabling Effective Identification Of Both Known And Unknown Intrusion Patterns. The System Pre-processes Captured Traffic, Extracts Relevant Features, And Performs Real-time Classification Of Network Behavior In To Normalormalicious Categories. Continuous Monitoring Ensures Immediate Detection And Response To Potential Cyber Threats, Thereby Enhancing Overall Network Resilience. Performance Evaluation Is Conducted Using Standard Metrics Such As Accuracy, Precision, Recall, And F1-score, Demonstrating Strong Detection Capability And Reliability. The Outcomes Indicate That The Proposed Research Offers A Scalable, Adaptive, And Efficient Cyber Security Framework Suitable For Deployment In Critical Domains Such As Finance, Healthcare, Cloud Environments, And Enterprise Networks, Where Real-time Protection And Intelligent Threat Analysis Are Essential. Read more Read less

TalentIQ-AN INTELLIGENT RESUME AND FAIR CANDIDATE RANKING SYSTEM

TalentIQ Is An AI-powered Resume Intelligence System Designed To Automate Candidate Screening And Improve Recruitment Efficiency. The System Analyzes Resumes Using Natural Language Processing (NLP) And Semantic Skill Matching Techniques To Identify T... TalentIQ Is An AI-powered Resume Intelligence System Designed To Automate Candidate Screening And Improve Recruitment Efficiency. The System Analyzes Resumes Using Natural Language Processing (NLP) And Semantic Skill Matching Techniques To Identify The Most Relevant Candidates. Unlike Traditional Applicant Tracking Systems (ATS) That Rely Solely On Keyword Matching, TalentIQ Incorporates Fraud Detection Mechanisms To Identify Exaggerated Or Misleading Resume Claims, Including Timeline Inconsistencies And Unsupported Skill Claims. The Framework Applies Anonymization Techniques To Remove Personally Identifiable Information, Thereby Reducing Bias And Ensuring Fair Candidate Evaluation During The Screening Process. An Explainable Ranking Mechanism Provides Transparent Justifications For Candidate Scores, Enabling Data-driven Hiring Decisions. The System Accepts Resumes In PDF And DOCX Formats, Performs Multi-factor Evaluation Combining Skill Relevance, Experience Validation, Authenticity Verification, And Fraud Risk Indicators, And Generates Ranked Candidate Leaderboards With Detailed Reasoning For Each Evaluation Decision. Read more Read less

GREEN SYNTHESIS AND CHARACTERIZATION OF COPPER OXIDE NANO PARTICLES USING SOLANUM NIGRUM LEAF EXTRACT

Green Synthesis Of Metal Nanoparticles Was Pacified In Its Beginning. Owing To The Enlarging Call Far Diverse Nanoparticles, It Is Inevitable To Evolve The Synthesis Manners That Are Advantageous And Ambiance Friendly. To Be Ascribed To The Eminent C... Green Synthesis Of Metal Nanoparticles Was Pacified In Its Beginning. Owing To The Enlarging Call Far Diverse Nanoparticles, It Is Inevitable To Evolve The Synthesis Manners That Are Advantageous And Ambiance Friendly. To Be Ascribed To The Eminent Cost And The Quality Of Physical And Chemical Approaches, The Necessity For Green Synthesis Of Nanoparticles Is Moderately Enlarging. On The Account Of Examine Of Low-priced Choices, Examiners Have Originate To Utilize The Biological Elements And Molecules That Act As Reducing Agents, Containing Micro-organisms, Bio- Molecules, Extracts From Plants, Moderately Enlarging. Copper Oxide Nanoparticles Was Synthesized By Using Solanum Nigrum Leaf Extract And Characterized By Following Methods. UV-Vis Absorption, XRD, EDX, FTIR And SEM Analyses Were Used To Characterize The Synthesized Copper Oxide Nanoparticles. Read more Read less

Intelligent System For Unified Medical History Tracking And Government Healthcare Scheme Eligibility Analysis

The Contemporary Healthcare Ecosystem Continues To Grapple With One Of Its Most Persistent Structural Problems: The Absence Of A Unified Platform Through Which Patient Data Can Flow Securely And Intelligently Across Care Providers. Fragmented Record-... The Contemporary Healthcare Ecosystem Continues To Grapple With One Of Its Most Persistent Structural Problems: The Absence Of A Unified Platform Through Which Patient Data Can Flow Securely And Intelligently Across Care Providers. Fragmented Record-keeping Forces Patients To Carry Physical Documents, Leads Physicians To Repeat Expensive Diagnostic Tests, And Prevents Timely Access To Full Clinical Histories. Against This Backdrop, This Paper Introduces HealthShield AI—a Cloud-native, AI-powered Web Platform That Consolidates Patient Medical Data Under A Universally Unique Health ID, Enabling Real-time, Role-controlled Access By Authorized Healthcare Providers At Any Affiliated Institution. Beyond Record Management, The System Incorporates An Intelligent Eligibility Engine That Matches Patients To Suitable Government-sponsored Health Schemes, A Specialist Centre Discovery Module, And An Interactive Shield AI Chatbot For Guided Decision Support. The System Is Implemented Using React.js, Node.js, Express.js, Supabase, And The OpenAI API. Evaluation On A Simulated Clinical Dataset Shows A Scheme Recommendation Accuracy Of 91.4%, An Average Record Retrieval Latency Of 340 ms, And A System Usability Scale Score Of 78.6, Collectively Affirming The Platform’s Clinical Utility And Readiness For Real-world Deployment. Read more Read less

AI-POWERED INDIAN MEDICINAL PLANT IDENTIFICATION AND INFORMATION SYSTEM USING DEEP LEARNING AND COMPUTER VISION TECHNOLOGY

Good Identification Of Medicinal Plants Is The Key To Maintaining The Traditional Knowledge Systems And Safe Use Of Herbs. Physical Identification Of Plants Is Time Consuming And Subject To Errors Especially Where Species Have The Same Morphology. Th... Good Identification Of Medicinal Plants Is The Key To Maintaining The Traditional Knowledge Systems And Safe Use Of Herbs. Physical Identification Of Plants Is Time Consuming And Subject To Errors Especially Where Species Have The Same Morphology. The Present Paper Describes An AI-based Indian Medicinal Plant Identification And Information System That Combines Convolutional Neural Networks With A Knowledge Retrieval System To Allow Real-time Identification Of Plants And Provide Surrounding Information About It. The Model Was Trained And Evaluated On A Curated Dataset Of 3,000 Images Of 50 Indian Medicinal Plant Species. Transfer Learning Was Used To Implement Two Deep Learning Architectures, ResNet18, And EfficientNet-B0. The Experimental Data Show That EfficientNet-B0 Was Able To Attain An Accuracy Of 90.3 Percent With A Latency Of 175 Ms To Run Inference, Surpassing Both ResNet18 And The Available Identification Systems (LeafSnap And PlantNet) To Identify Plants In Controlled Conditions. The System Also Includes Knowledge Base With Structured Knowledge Base And Language Model-based Retrieval System To Produce Detailed Information Such As Medicinal Uses, Phytochemical Properties, And Precautionary Advice To The Identified Species. Application Deployment Via A Streamlit Interface And Containerized Cloud Hosting Also Allows It To Scale And Be Accessed In Real Time. The Suggested Framework Illustrates The Usage Of The Computer Vision And Language-based Intelligence In An Efficient Manner, To Aid In The Field-level Medicinal Plant Detection And Digital Conservation Of The Herbal Knowledge. Read more Read less

AI-Based Surveillance System For Abandoned Object Detection Using YOLOv8

The Proliferation Of Surveillance Infrastructure Across Transportation Hubs, Commercial Complexes, And Civic Spaces Has Not Been Matched By A Proportional Improvement In The Capacity Of Human Operators To Effectively Monitor Multiple Video Feeds Over... The Proliferation Of Surveillance Infrastructure Across Transportation Hubs, Commercial Complexes, And Civic Spaces Has Not Been Matched By A Proportional Improvement In The Capacity Of Human Operators To Effectively Monitor Multiple Video Feeds Over Extended Durations. Cognitive Limitations Inherent To Sustained Visual Monitoring Create Critical Gaps During Which Security-relevant Events, Including The Placement Of Unattended Items, May Go Unnoticed. This Paper Proposes A Computationally Efficient, Learning-based Framework That Autonomously Identifies Abandoned Personal Belongings Within Live Surveillance Footage By Integrating The YOLOv8 Single-stage Detector With A Spatiotemporal Ownership Inference Mechanism. The System Processes Individual Video Frames To Simultaneously Detect Persons And Personal Items — Encompassing Backpacks, Handbags, Laptops, And Mobile Devices — Using COCO-pretrained Weights Applied In A Zero-shot Configuration. Temporal Continuity Is Preserved Through An IoU Matching Strategy And Ownership Attribution Is Determined By Euclidean Proximity Analysis. Should The Attended Condition Remain Unsatisfied Beyond A User-defined Temporal Threshold, The Item Is Reclassified As Abandoned, Prompting A Visual Alert. Experimental Validation Confirms Near-real-time Throughput And Reliable Detection Outcomes. Read more Read less

Cyber Threat Intelligence System Using CNN-LSTM Deep Learning Model

Cyber Threat IntelligenceThe Rapid Evolution Of Cyber Threats, Including Malware, Phishing Attacks, Distributed Denial Of Service (DDoS), And Advanced Persistent Threats, Has Made Traditional Security Mechanisms Less Effective. Conventional Rule-base... Cyber Threat IntelligenceThe Rapid Evolution Of Cyber Threats, Including Malware, Phishing Attacks, Distributed Denial Of Service (DDoS), And Advanced Persistent Threats, Has Made Traditional Security Mechanisms Less Effective. Conventional Rule-based And Signature-based Detection Systems Fail To Identify Unknown Or Zero-day Attacks, As They Rely Heavily On Predefined Patterns. This Limitation Highlights The Need For Intelligent And Adaptive Cyber Threat Intelligence (CTI) Systems Capable Of Analyzing Large-scale And Complex Data In Real Time. This Project Proposes A Cyber Threat Intelligence System Using A Hybrid CNN-LSTM Deep Learning Model To Enhance Threat Detection And Classification. The System Combines The Strengths Of Convolutional Neural Networks (CNN) And Long Short-Term Memory (LSTM) Networks. CNN Is Utilized For Efficient Feature Extraction From High-dimensional Network Traffic Data, Identifying Important Spatial Patterns Such As Anomalies And Malicious Signatures. LSTM, A Type Of Recurrent Neural Network (RNN), Is Employed To Capture Temporal Dependencies And Sequential Patterns In The Data, Which Are Essential For Detecting Evolving And Time-based Cyber Attacks. The System Processes Large Cyber Security Datasets, Such As Network Traffic Logs, By Performing Data Preprocessing Steps Including Normalization, Noise Removal, And Feature Transformation. The Processed Data Is Then Fed Into The CNN-LSTM Model For Training And Prediction. The Model Classifies Network Activities Into Multiple Categories Such As Normal Traffic, Intrusion Attempts, Malware, And DDoS Attacks. Experimental Results Demonstrate That The Proposed Hybrid Model Achieves Higher Accuracy, Better Precision, And Reduced False Positive Rates Compared To Traditional Machine Learning Models Like Support Vector Machines (SVM) And Decision Trees. Additionally, The System Is Capable Of Handling Large-scale Data Efficiently, Making It Suitable For Real-time Threat Detection Environments. Read more Read less

A PRIVACY-ENHANCED CRYPTO-BIOMETRIC AUTHENTICATION FRAMEWORK

Airport Authority Systems Demand The Highest Levels Of Security For Controlling Access To Restricted Zones, Yet Traditional Authentication Mechanisms Based On Identity Cards, PINs, And Passwords Remain Inherently Susceptible To Theft, Forgery, Duplic... Airport Authority Systems Demand The Highest Levels Of Security For Controlling Access To Restricted Zones, Yet Traditional Authentication Mechanisms Based On Identity Cards, PINs, And Passwords Remain Inherently Susceptible To Theft, Forgery, Duplication, And Social Engineering Attacks. Biometric Authentication Offers A More Reliable Alternative; However, Conventional Face Recognition Systems Face Critical Challenges Including Spoofing Vulnerabilities, Inadequate Privacy Preservation For Stored Biometric Templates, And Limited Accuracy In Live-face Detection. This Paper Proposes A Privacy-Enhanced Crypto-Biometric Authentication Framework (PECBAF) Specifically Tailored For Airport Authority Access Control, Extending The Cancellable Template Protection Scheme Of Imran Et Al. [1] To The Facial Biometric Domain. The Proposed Framework Integrates The Grassmann Manifold Algorithm For Live Facial Feature Extraction, Deriving Approximately 1024 Discriminative Features From 76 Key Facial Landmark Points. The Extracted Facial Feature Subspace Representation Is Subjected To A Möbius Conformal Transformation Parameterized By A User-specific Keyset (p, Q), Generating A Non-invertible Cancellable Biometric Template That Cannot Be Reconstructed From Stored Data Even Under Database Compromise. The Cancellable Template Is Subsequently Secured Using A Dual-layer Hybrid Encryption Scheme Combining 256-bit AES With RSA Public-key Cryptography, Safeguarding Template Integrity Against Spoofing, Brute-force, And Cross-database Attacks. An Access Control Module Enforces Role-based Entry Decisions With Real-time Audit Logging. The Proposed System Is Evaluated In Terms Of Revocability, Unlinkability, Non-invertibility, And Authentication Accuracy. Results Demonstrate High Genuine Acceptance Rates With Low False Acceptance Rates, Confirming The System's Suitability For High-security Airport Environments. The Framework Provides A Robust, Privacy-preserving, And Computationally Efficient Alternative To Conventional Airport Biometric Systems. Read more Read less

IOT DIGITAL TWIN BASED AIR QUALITY MONITORING SYSTEM

In Recent Times IoT (Internet Of Thing) Plays A Major Role In The Daily Routine Of Human Life. Majorly This Technology Is Showing Its Impact In Medical Applications. In Contrast, Micro Stations As A Kind Of Low-cost Air Monitoring Equipment Can Be Di... In Recent Times IoT (Internet Of Thing) Plays A Major Role In The Daily Routine Of Human Life. Majorly This Technology Is Showing Its Impact In Medical Applications. In Contrast, Micro Stations As A Kind Of Low-cost Air Monitoring Equipment Can Be Distributed Densely Though Their Accuracy Is Relatively Low. This Paper Proposes A Deep Calibration Method For Low-cost Air Monitoring Sensors Equipped In The Micro Stations, Which Consists Of Two Sensors For CO2 And PM2.5 That Are Important For Air Quality Monitoring With Compensated Weather Monitoring Capabilities Were Deployed In The Villages. The End Users Can Query The System And Access The Data Together With The Analytic Information Via The Developed Web-based User Interface Dashboard. Baseline Algorithms Called Predictive Method That Facilitate Setting Of Triggers For Each Sensing Node And Pushing Of Notifications For When A Measured Parameter Exceeds A Certain Threshold Value Are Proposed And Implemented. Later, An IoT Based Monitoring System Has Been Developed To Measure The Air Quality But Monitoring And Gathering The Data From Multiple Nodes. Read more Read less

“Trashbot” The Smart Dustbin

As Global Urban Populations Expand, The Inefficiencies Of Traditional, Stationary Waste Management Systems Have Become Increasingly Apparent, Often Resulting In Environmental Degradation And Public Health Risks. To Address These Limitations, This Res... As Global Urban Populations Expand, The Inefficiencies Of Traditional, Stationary Waste Management Systems Have Become Increasingly Apparent, Often Resulting In Environmental Degradation And Public Health Risks. To Address These Limitations, This Research Introduces TrashBot, An Autonomous, Omnidirectional Robotic Waste Collection System That Leverages Artificial Intelligence And Computer Vision To Transform Waste Disposal From A Passive To An Active Process. The System Utilizes The YOLO (You Only Look Once) Deep Learning Framework For Real-time Trash Identification Across Diverse Environments. A Key Innovation Of TrashBot Is Its Predictive Interception Capability; By Applying Projectile Motion Analysis, The Robot Calculates The Trajectory Of Discarded Items To Proactively Position Itself At The Estimated Landing Point. Mobility Is Facilitated By An Omnidirectional Drivetrain, Allowing For Rapid, Multi-directional Maneuvers In Crowded Spaces. For Safe Navigation, The System Integrates Ultrasonic-based Environmental Mapping And The A Path-planning Algorithm* To Navigate Dynamic Obstacles Efficiently. By Synthesizing Advancements In Autonomous Robotics And Predictive Modeling, TrashBot Offers A Scalable Solution For Modern Waste Management That Reduces Manual Labor And Improves Urban Hygiene. The Modular Architecture Further Allows For Future Integrations, Such As Automated Sorting And Sanitization, Aligning With The Broader Objectives Of Developing Sustainable, Smart City Infrastructures. Read more Read less

CONSEQUENCE INTELLIGENCE ENGINE FOR PREDICTING DECISION IMPACT IN COMPLEX SYSTEMS

Strategic Decision-making In Complex Organizational Environments Is Inherently Associated With Risk And Uncertainty. Despite Widespread Adoption Of Decision Support Systems, Most Existing Tools Focus On Structuring Decisions Rather Than Predicting Th... Strategic Decision-making In Complex Organizational Environments Is Inherently Associated With Risk And Uncertainty. Despite Widespread Adoption Of Decision Support Systems, Most Existing Tools Focus On Structuring Decisions Rather Than Predicting Their Downstream Consequences. This Limitation Leaves Organizations Without The Intelligence Required To Anticipate Risk Trajectories, Quantify Impact Severity, Or Understand The Temporal Evolution Of A Decision's Effects. The Consequence Intelligence Engine (CIE) Addresses This Gap By Introducing A Machine Learning-driven Framework Capable Of Predicting The Risk And Organizational Impact Of Strategic Decisions Before Execution. The Proposed System Accepts Natural Language Decision Descriptions Combined With Structured Organizational Context Parameters, Including Company Size, Industry Type, Market Condition, Growth Stage, Workforce Morale, Attrition Trends, And Risk Appetite. A Random Forest Regressor Is Employed As The Core Prediction Model, Trained On A Synthetic Scenario-based Dataset Capturing Diverse Organizational Decision Profiles. Feature Engineering And Context-aware Encoding Enable The Model To Generate An Overall Risk Score, Categorized Risk Classifications, And A Temporal Impact Timeline Spanning Short-term Through Long-term Horizons. A FastAPI Backend Exposes The Prediction Engine As A RESTful Service, While A React-based Interactive Dashboard Presents Results In An Interpretable And Actionable Format. The System Contributes A Novel Approach To Decision Intelligence By Combining Predictive Modeling, Temporal Risk Analysis, And Contextual Awareness In A Unified, Accessible Platform. Read more Read less

CLUSTBIGFIM: MAPREDUCE CF FOR BIG DATA ITEMSET MINING

Frequent Itemset Mining (FIM) Is Essential For Discovering Patterns In Large-scale Data, But Traditional Algorithms Struggle With Big Data Volumes Due To Scalability Issues. ClustBigFIM Introduces A Hybrid MapReduce-based Framework That Integrates Pa... Frequent Itemset Mining (FIM) Is Essential For Discovering Patterns In Large-scale Data, But Traditional Algorithms Struggle With Big Data Volumes Due To Scalability Issues. ClustBigFIM Introduces A Hybrid MapReduce-based Framework That Integrates Parallel K-means Clustering As Preprocessing To Partition Datasets Into Manageable Clusters, Followed By Modified BigFIM Employing Apriori And Eclat Algorithms For Efficient Extraction Of Frequent Itemsets. In The MapReduce Paradigm, The Map Phase Computes Distances And Assigns Itemsets To Clusters, While The Reduce Phase Aggregates Results And Generates Patterns Useful For Business Analytics Like Market Basket Analysis. Evaluated On Large Synthetic And Real-world Datasets, ClustBigFIM Achieves Superior Speedup, Scalability, And Execution Time Compared To Standalone BigFIM By Reducing Data Redundancy Through Clustering. This Approach Leverages Hadoop’s Fault-tolerant Processing To Handle Petabyte-scale Data, Enabling Robust FIM In Distributed Environments. Read more Read less

Hybrid YOLOv8–CNN Framework For Automated Analysis Of Temporal FMRI Brain Networks

Functional Magnetic Resonance Imaging (fMRI) Provides Critical Insights Into Dynamic Brain Activity And Connectivity Patterns, Enabling The Study Of Neural Functions Over Time. Traditional Clustering Methods, Such As Topological Data Analysis (TDA),... Functional Magnetic Resonance Imaging (fMRI) Provides Critical Insights Into Dynamic Brain Activity And Connectivity Patterns, Enabling The Study Of Neural Functions Over Time. Traditional Clustering Methods, Such As Topological Data Analysis (TDA), Are Limited To Grouping Brain Networks And Cannot Accurately Detect Or Classify Active Regions Or Abnormalities Like Tumors. Manual Interpretation Of MRI And FMRI Data Is Time-consuming, Prone To Human Error, And Often Inconsistent Across Multi-site Datasets. To Overcome These Challenges, The Proposed System Introduces A Hybrid YOLOv8 And Convolutional Neural Network (CNN) Framework For Automated Detection, Localization, Classification, And Staging Of Brain Tumors Along With Analysis Of Normal Brain Activity. YOLOv8 Precisely Detects And Localizes Active Brain Regions In MRI And FMRI-derived Maps, Generating Bounding Boxes And Confidence Scores. The CNN Extracts Deep Spatial And Temporal Features Enabling Accurate Classification Of Brain States And Tumor Stage Determination. Experimental Results Demonstrate High Accuracy In Brain Activity Analysis, Tumor Detection, And Staging. This Automated Method Reduces Dependence On Manual Interpretation, Providing Faster, More Reliable, And Interpretable Clinical Insights. Read more Read less

Smart Cybersecurity Intrusion Detection & Prevention System Using AI (SecurityHub)

This Project Examines The Combination Of Network-level Security And Physical Biometric Verification Within A Unified, Hardware-accelerated Edge Environment. Conventional Security Setups Frequently Struggle With Operational Slowdowns, Primarily Caused... This Project Examines The Combination Of Network-level Security And Physical Biometric Verification Within A Unified, Hardware-accelerated Edge Environment. Conventional Security Setups Frequently Struggle With Operational Slowdowns, Primarily Caused By High-latency Cloud Dependencies And The Large Bandwidth Needs Of Remote Data Handling. To Fix These Issues, We Present A Localized Security Hub Designed For The Raspberry Pi 5 And The Hailo-8L AI Accelerator. This Setup Allows For The Simultaneous Running Of Deep Packet Inspection Through Suricata And Real-time Facial Recognition Using The ArcFace Model. By Integrating Keycloak As A Centralized Identity Provider, The System Ensures A Unified Security Environment Where Network Access Rules Are Assigned Based On Biometrically Verified Roles. This Research Shows That Enterprise-level Security—capable Of Managing Both Digital Threats And Physical Breaches—is Possible On A Low-power, Single-board Device. Our Results Show That Using The Dedicated PCIe Gen 3 Interface On Raspberry Pi 5 For AI Tasks Significantly Improves Speed While Protecting Data Privacy. Ultimately, This Method Offers A Sturdy, Small Scale Budget-friendly Alternative To Traditional Isolated And Expensive Security Tools, Creating A Reliable Zero-Trust Environment At The Network Edge. Read more Read less

Assessment Of RCC T-Beam Bridge Superstructure Under Different Codes And Loading Conditions

This Study Summarized Comparative Design And Analysis Of RCC T-Beam Bridge Superstructure For Different Codes I.e.,Indian Road Congress (IRC) Codes And American Association Of State Highway Transportation Officials (AASHTO) Specification Load Combina... This Study Summarized Comparative Design And Analysis Of RCC T-Beam Bridge Superstructure For Different Codes I.e.,Indian Road Congress (IRC) Codes And American Association Of State Highway Transportation Officials (AASHTO) Specification Load Combinations For Varying Span Length. The Several Codes Are Used To Design The Bridges. IRC 21-2000 Used For Designing Bridges By Working Stress Method (WSM), Also IRC: 112-2011 Introduced By Indian Road Congress For RCC And Pre- Stressed Bridges By Limit State Method (LSM). Both The Codes Have Different Guidelines And Procedure For Design Of Bridges. This Study Based On IRC 112-2011 (LSM) And IRC: 6- 2017 Is Used For Load Considerations. In Which This Analysis Depends On The Analytical Modelling By Finite Element Method (FEM) For In STAAD-Pro Software And Comparing The Structural Parameter Bending Moment, Shear Force, Deflection And Area Of Reinforcement For Different Girder Span Length 16M, 20M, 24M As Per The IRC And AASHTO Code. Class A & Class 70R Consider From IRC 6-2017 And HS93 Is The Vehicular Loading Consider From AASHTO. Form The Analysis Understanding Suitability Design Technique And The Behavior Of Two-lane Carriage Way Width Of T- Beam Bridge Superstructure Under Different Loading Condition And By Using Different Code And Comparing The Result, Conclusions Will Be Made That Up To What Extents Similarities Between Both Standards. Read more Read less

Literature Review On Howe Truss Bridge Using Hollow And Open Steel Section By Using STAAD Pro

A Howe Truss Bridge Is A Widely Used Structural Configuration Known For Its Efficiency In Load Distribution And Suitability For Medium To Long Spans. This Study Presents A Comparative Analysis Of A Howe Truss Bridge Using Hollow Steel Sections And Op... A Howe Truss Bridge Is A Widely Used Structural Configuration Known For Its Efficiency In Load Distribution And Suitability For Medium To Long Spans. This Study Presents A Comparative Analysis Of A Howe Truss Bridge Using Hollow Steel Sections And Open Steel Sections, Modeled And Analyzed With STAAD Pro. The Primary Objective Is To Evaluate The Structural Performance, Material Efficiency, And Economic Feasibility Of Both Section Types Under Identical Loading And Boundary Conditions. The Bridge Models Are Subjected To Dead Load, Live Load, And Environmental Loads In Accordance With Relevant Design Standards. Key Parameters Such As Deflection, Stress Distribution, Weight Of Structure, And Factor Of Safety Are Assessed. Hollow Steel Sections, Due To Their Closed Geometry, Exhibit Higher Torsional Rigidity And Improved Resistance To Buckling, Whereas Open Steel Sections Offer Ease Of Fabrication And Lower Initial Cost. The Analysis Results Highlight The Advantages And Limitations Of Each Section Type In Terms Of Strength, Stability, And Overall Performance. The Study Concludes By Identifying The More Efficient Section Type For Howe Truss Bridges Based On Structural Behavior And Cost-effectiveness, Providing Useful Insights For Engineers In The Design And Optimization Of Truss Bridge Systems. Read more Read less

LLM-Based Medical AI Chatbot Using LLaMA For Healthcare Assistance

Access To Quality Healthcare Remains A Critical Global Challenge. Approximately 3.5 Billion People Lack Access To Basic Healthcare Services. This Paper Presents The Design And Development Of A Full-stack Document Intelligence Application Powered By A... Access To Quality Healthcare Remains A Critical Global Challenge. Approximately 3.5 Billion People Lack Access To Basic Healthcare Services. This Paper Presents The Design And Development Of A Full-stack Document Intelligence Application Powered By Architecture For Intelligent Healthcare Query Support. The System Allows Users To Upload Documents (PDF, DOCX, TXT), Which Are Semantically Chunked, Embedded Using Google Generative AI Embedding, And Stored In A FAISS Vector Store. Upon Receiving A User Query, Relevant Documents Are Retrieved From FAISS And Passed As Context To The Gemini-2.0-Flash Model, Which Generates Accurate, Hallucination-reduced Responses Grounded In Verified Medical Literature. The Proposed Architecture Incorporates A Curated Medical Knowledge Base Scraped From WebMD (1,613 Health Topics, 27,744 Chunks), Chain-of-thought Prompt Engineering, Session-aware Chat History Management, And A Multimodal Image-based Retrieval Pipeline Using Gemini Flash 2.0. The System Integrates Textual Prompt Analysis And Medical Image Interpretation, Supporting Preliminary Medical Diagnosis For Underserved Communities. Experimental Evaluation Across 30 Conversation Sessions And 119 Medical Queries Demonstrates An Average Response Time Of 3.6 Seconds, Relevance 4.8/5, Fluency 4.9/5, And User Safety 4.2/5, Substantially Outperforming Standalone LLM Querying. Read more Read less

Tamper-Proof Supply Chain System For Counterfeit Detection Using Blockchain

This System Presents A Blockchain Based Secure Product Tracking And Authenticity Verification Solution For Modern Supply Chains. It Enables Tracking Of Products At Every Stage, From Manufacturer To End User, Ensuring Complete Transparency. Blockchain... This System Presents A Blockchain Based Secure Product Tracking And Authenticity Verification Solution For Modern Supply Chains. It Enables Tracking Of Products At Every Stage, From Manufacturer To End User, Ensuring Complete Transparency. Blockchain Technology Is Used To Store All Product Data In A Decentralized And Tamper-proof Manner. Each Product Is Assigned A Unique QR Code Or Barcode That Links Directly To Its Blockchain Record. This Allows Consumers To Verify Product Authenticity In Real Time. The System Uses The SHA-256 Hashing Algorithm To Securely Store And Validate Product Information. This Ensures Data Integrity And Prevents Unauthorized Modifications Throughout The Product Lifecycle. Every Transaction In The Supply Chain Is Recorded And Cannot Be Altered Once Stored. The System Reduces Dependency On Centralized Databases And Increases Reliability. It Significantly Minimizes The Risk Of Counterfeit Products Entering The Market. It Also Improves Trust And Traceability Among All Stakeholders. Overall, The System Enhances Security, Transparency, And Efficiency For Manufacturers, Distributors, And Consumers. Read more Read less

Detection of Diabetic Retinopathy Using a Multi-Decision Inception-ResNet Blended Hybrid Model

Diabetic Retinopathy (DR) Represents A Critical Complication Of Diabetes Mellitus, Leading To Progressive Vision Impairment And Potential Blindness If Left Undetected. This Research Presents A Novel Multi-decision Inception-ResNet Blended Hybrid Mode... Diabetic Retinopathy (DR) Represents A Critical Complication Of Diabetes Mellitus, Leading To Progressive Vision Impairment And Potential Blindness If Left Undetected. This Research Presents A Novel Multi-decision Inception-ResNet Blended Hybrid Model For Automated DR Detection And Classification. The Proposed Architecture Integrates 172 Weighted Layers, Strategically Divided Into Dual-image Processing Pathways: 86 Layers Dedicated To Color Fundus Image Analysis And 86 Layers For Grayscale Image Processing. By Employing A Multi-layered Transfer Learning Approach With Adaptive Moment Estimation (Adam) And Stochastic Gradient Descent (SGD) Optimization Techniques, The Model Achieves Comprehensive Feature Extraction Across Both Sequential And Non-sequential Image Data. The Architecture Incorporates Eight Convolutional Layers At Each Processing Stage, Enabling The Extraction Of Both Global And Specialized Features Through Chi-square Testing Mechanisms. Evaluated On The EyePACS And APTOS Datasets, The Model Demonstrates Superior Performance With A Detection Accuracy Of 98.1%, Outperforming Existing State-of-the-art Approaches. The Multi-decision Framework Effectively Classifies DR Into Five Severity Stages: No DR, Mild DR, Moderate DR, Severe DR, And Proliferative DR, Providing A Robust Solution For Early-stage Diabetic Retinopathy Detection In Clinical Settings. Read more Read less

Structural Analysis And Feasibility Design Of Pervious Baffle Walls In Sewer For Wastewater Treatment

This Research Investigates The Structural Analysis And Feasibility Of Foam-based Pervious Baffle Walls For Wastewater Treatment Applications. Conventional Baffle Walls Are Impermeable And Subjected To High Hydrostatic Pressure, Leading To Cracking An... This Research Investigates The Structural Analysis And Feasibility Of Foam-based Pervious Baffle Walls For Wastewater Treatment Applications. Conventional Baffle Walls Are Impermeable And Subjected To High Hydrostatic Pressure, Leading To Cracking And Durability Issues. The Proposed System Utilizes Pervious Concrete Incorporating Foam And Fly Ash To Reduce Self-weight And Allow Controlled Water Flow. Experimental And Analytical Studies Were Conducted To Evaluate Compressive Strength, Permeability, And Structural Performance. The Results Demonstrate That The System Provides An Effective Balance Between Strength And Permeability, Making It A Sustainable And Cost-effective Solution For Sewer Infrastructure. If Demand Of Energy Is Higher, Then It Will Generate The Electricity In Thermal Power Plant And This Power Plant Having Waste Material Like Fly Ash. With Rapid Growth Of Industrial Development With Large Demand Of Electricity Has Led To 100 Million Tones Fly Ash (around 70% Of Fly Ash) Being Discharged Every Year Worldwide. In This Project Research, We Used This Fly Ash To Create The Wastewater Concrete Blocks To Purify The Wastewater And In Making Of This Concrete Block. Read more Read less

An Intelligent Context-Aware Web Browser For Assisted Research

Modern Web Browsers Are Optimized For Content Navigation And Rendering Rather Than Information Understanding. Researchers And Students Often Rely On Multiple Browser Tabs To Gather Information, Which Leads To Cognitive Overload, Fragmented Context, A... Modern Web Browsers Are Optimized For Content Navigation And Rendering Rather Than Information Understanding. Researchers And Students Often Rely On Multiple Browser Tabs To Gather Information, Which Leads To Cognitive Overload, Fragmented Context, And Reduced Productivity. Although Recent AI-powered Tools Provide Assistance In Summarization And Question Answering, They Typically Function As External Add-ons And Lack Continuous Awareness Of User Intent And Browsing Context. This Paper Proposes An Intelligent, Context-aware Web Browser Framework That Actively Assists Users During Research Activities By Understanding Content Semantics, Tracking User Intent, And Providing Real-time Insights Such As Summaries, Comparisons, And Knowledge Connections. The Proposed Approach Aims To Transform Conventional Browsers From Passive Interfaces Into Proactive Research Assistants, Thereby Improving Research Efficiency And User Experience. Read more Read less

A Novel Steganographic Approach To Strengthen Enhanced MFA And Attack Prevention For Credential Transmission

Digital Banking Infrastructure Faces Escalating And Sophisticated Threats Including Phishing, Man-in-the-Middle (MITM) Interceptions, Session Hijacking, Replay Attacks, Credential Stuffing, And Denial-of-service (DoS) Campaigns. Conventional Single-f... Digital Banking Infrastructure Faces Escalating And Sophisticated Threats Including Phishing, Man-in-the-Middle (MITM) Interceptions, Session Hijacking, Replay Attacks, Credential Stuffing, And Denial-of-service (DoS) Campaigns. Conventional Single-factor Authentication Mechanisms Based On Username-password Pairs Offer Insufficient Protection, While Existing Multi-Factor Authentication (MFA) Implementations—such As SMS-based One-Time Passwords (OTP), Hardware Tokens, And Basic Biometric Checks—continue To Exhibit Exploitable Vulnerabilities. This Paper Proposes A Novel Five-layer Secure Authentication And Transaction Authorization Framework Tailored To Digital Banking Environments. The System Integrates: (i) Grassmann Manifold-based Facial Recognition For Biometric Enrollment And Live Verification, Replacing Hardware USB Tokens With A Mathematically Robust Biometric Factor; (ii) Multi-factor Login Combining Credential-based Authentication With Biometric Matching; (iii) Dynamic Per-session Cryptographic Key Generation Using SHA-512 With User-specific Salts; (iv) QR-code Least Significant Bit (LSB) Steganography For Covert Session Key Transmission To The User's Registered Email, Hiding Sensitive Token Data Within An Innocuous Carrier Image; And (v) Per-transaction Session Key Validation With Real-time Unauthorized-access Alerting. The Proposed Architecture Extends And Improves Upon The Secure Multi-Factor Authentication (SMFA) Framework By Sarower Et Al. [1] By Eliminating Physical Device Dependency, Adding Biometric Security, And Introducing A Banking-domain-specific Steganographic Session Key Channel. Security Analysis Via Burrows-Abadi-Needham (BAN) Logic Demonstrates Protocol Correctness. The Facial Recognition Module Achieves A 97.3% True Acceptance Rate (TAR) With A False Acceptance Rate (FAR) Below 0.8%. Steganographic Embedding Achieves A PSNR Of 43.2 DB, Well Above The 40 DB Imperceptibility Threshold. Total Authentication Pipeline Latency Is Approximately 2.3 Seconds On Standard Hardware. Read more Read less

Literature Review On Structural And Hydraulic Analysis And Design Of Retaining Walls For Flood Mitigation

Floods Are Increasingly Frequent And Destructive Due To Urbanization, Climate Change, And Unpredictable Rainfall, Making Effective Flood Management Essential. Retaining Walls In Flood-prone Areas Must Resist Not Only Earth Pressure But Also Hydrostat... Floods Are Increasingly Frequent And Destructive Due To Urbanization, Climate Change, And Unpredictable Rainfall, Making Effective Flood Management Essential. Retaining Walls In Flood-prone Areas Must Resist Not Only Earth Pressure But Also Hydrostatic, Dynamic, And Seepage Forces That Are Often Neglected In Conventional Designs. This Study Focuses On The Integrated Structural And Hydraulic Design Of Retaining Walls To Ensure Stability Against Sliding, Overturning, And Foundation Failure Under Combined Loading Conditions. It Considers Key Factors Such As Soil Properties, Wall Geometry, Drainage Systems, And Surcharge Loads, Along With The Use Of Advanced Tools Like STAAD.Pro For Accurate Analysis. The Outcome Provides Efficient, Safe, And Sustainable Design Solutions For Retaining Walls To Enhance Resilience And Protect Infrastructure In Flood-affected Regions. Read more Read less

STABILIZATION OF EXPANSIVE SOIL USING STONE DUST AND FLY ASH: A REGIONAL STUDY OF SOUTH GUJARAT

Expansive (black Cotton) Soils Pose Serious Challenges To Civil Engineering Structures Due To Their High Swelling And Shrinkage Behavior. This Study Focuses On Stabilizing Such Soils Using Fly Ash And Stone Dust As Economical And Sustainable Material... Expansive (black Cotton) Soils Pose Serious Challenges To Civil Engineering Structures Due To Their High Swelling And Shrinkage Behavior. This Study Focuses On Stabilizing Such Soils Using Fly Ash And Stone Dust As Economical And Sustainable Materials. Various Proportions Of Fly Ash (5%, 10%, 15%) And Stone Dust (10%, 20%, 30%) Were Added To The Soil. Laboratory Tests Including Standard Proctor Test, Atterberg Limits, And Shear Strength Tests Were Conducted. The Results Showed An Increase In Maximum Dry Density From 1589.65 Kg/m³ To 1842.89 Kg/m³ And A Decrease In Optimum Moisture Content From 21.97% To 16.23%. The Liquid Limit And Plastic Limit Were Significantly Reduced, And Shear Strength Increased From 124.86 KPa To 412.28 KPa. The Study Concludes That The Optimum Mix Of 15% Fly Ash And 30% Stone Dust Effectively Improves Soil Properties, Making It Suitable For Construction Applications. Read more Read less

RETROFITTING OF BALANCE CANTILEVER NARMADA RIVER 1st BRIDGE THROUGH SLAB, PIER CAP AND BEARING REPLACEMENT

The Present Study Deals With The Retrofitting And Rehabilitation Of The First Balanced Cantilever Bridge Over The Narmada River Through Slab Replacement, Pier Cap Strengthening, And Bearing Replacement. Due To Aging, Increased Traffic Loading, And En... The Present Study Deals With The Retrofitting And Rehabilitation Of The First Balanced Cantilever Bridge Over The Narmada River Through Slab Replacement, Pier Cap Strengthening, And Bearing Replacement. Due To Aging, Increased Traffic Loading, And Environmental Effects, The Existing Bridge Components Showed Deterioration Affecting Structural Performance And Serviceability. To Restore The Bridge Capacity And Extend Its Service Life, A Detailed Structural Assessment And Retrofitting Methodology Were Carried Out. The Bridge Consists Of A Total Length Of 1360 M With 14 Spans Of 100 M Each. Structural Analysis Was Performed Using MIDAS Civil Software Considering Dead Load, Live Load, Wind Load, Braking Load, Temperature Effects, And Seismic Forces As Per Relevant IRC Provisions. Dynamic Analysis Using The Response Spectrum Method Was Also Conducted To Evaluate Seismic Behavior. The Proposed Retrofit Method Involved Replacement Of The Existing Slab And Bearings While Retaining The Foundation And Pier System. The Results Showed A Reduction Of Approximately 1000 Tons In Superstructure Dead Load, Which Significantly Reduced The Load On The Substructure And Improved Overall Bridge Performance. The Study Concludes That The Proposed Retrofitting Approach Is Economical, Safe, And Effective Compared To Complete Bridge Reconstruction, While Ensuring Enhanced Durability And Extended Service Life. Read more Read less

TweetSense: Emotion Detection From Twitter Data Using Natural Language Processing

This Research Investigates The Application Of Natural Language Processing (NLP) Models For Sentiment Analysis Of Twitter Data, A Domain That Has Gained Immense Importance In The Era Of Digital Communication.[1] Twitter, As A Microblogging Platform, G... This Research Investigates The Application Of Natural Language Processing (NLP) Models For Sentiment Analysis Of Twitter Data, A Domain That Has Gained Immense Importance In The Era Of Digital Communication.[1] Twitter, As A Microblogging Platform, Generates Millions Of Short Text Messages Daily, Reflecting Public Opinion On Diverse Subjects Ranging From Politics And Business To Entertainment And Social Issues. The Brevity And Informality Of Tweets, Combined With The Use Of Slang, Abbreviations, Emojis, And Multilingual Expressions, Make Sentiment Classification A Challenging Task.[2]The Proposed Framework Integrates Preprocessing Techniques, Feature Extraction Methods, And Advanced Machine Learning And Deep Learning Models To Classify Tweets Into Positive, Negative, Or Neutral Sentiments.[3] Transformer-based Architectures Such As BERT And RoBERTa Are Employed To Capture Contextual Meaning, Thereby Improving Classification Accuracy.[4] The System Is Designed To Provide Real-time Sentiment Monitoring, Enabling Stakeholders To Track Public Opinion Trends And Make Informed Decisions.[5][6] This Study Contributes To The Growing Field Of Social Media Analytics By Offering A Scalable, Efficient, And Accurate Solution For Sentiment Analysis.[7]- Read more Read less

GEO-FENCE INTEGRATED IoT FRAMEWORK FOR CHILDREN HEALTH AND SAFETY ASSURANCE

The Current Invention Is About A Geo-Fence Integrated Internet Of Things (IoT) Framework Focused On Health And Safety Insurance For Children. This System Combines Real-time Monitoring Of Health, Detection Of Environmental Hazards, And Location Tracki... The Current Invention Is About A Geo-Fence Integrated Internet Of Things (IoT) Framework Focused On Health And Safety Insurance For Children. This System Combines Real-time Monitoring Of Health, Detection Of Environmental Hazards, And Location Tracking Into A Single Embedded Platform. The Device Uses An ESP32 Microcontroller Along With A Pulse Sensor, Temperature Sensor, Smoke Sensor, Ultrasonic Sensor, GSM-GPS Module, And A Camera Module To Continuously Monitor A Child’s Health And Surroundings. The Framework Allows For Smart Monitoring By Gathering, Processing, And Sending Data Through Embedded C Programming. The System Tracks Essential Health Metrics Like Heart Rate And Body Temperature While Also Detecting Environmental Risks Such As Smoke And Unsafe Proximity To Hazards. The GSM-GPS Module Provides Real-time Location Tracking And Geo-fencing, Letting Users Set Up Safe Boundaries. If There Are Abnormal Health Readings, Environmental Dangers, Or Boundary Breaches, Automated Alerts Are Sent To Registered Guardians Using GSM Communication And Cloud Services. The Camera Module Boosts Monitoring By Offering Visual Confirmation In Emergencies. The Device Runs On A 12V Battery, Which Is Converted And Regulated To 5V DC To Maintain Stable And Safe Operation Of All Components. The System Is Built To Use Low Power, Ensure Reliable Data Transmission, And Maintain Continuous Monitoring. By Bringing Together Multiple Safety Features In A Compact IoT Framework, This Invention Offers An Effective And Affordable Solution For Monitoring Child Health, Detecting Environmental Safety Issues, And Managing Location-based Security. Read more Read less

GESTURE-CONTROLLED SMART GLOVE WITH VOICE ASSISTANCE AND ALERT SYSTEM

Communication Remains A Fundamental Human Need, Yet Individuals With Hearing And Speech Impairments Face Persistent Barriers In Daily Interactions. Traditional Methods Like Sign Language Require Mutual Proficiency Between Participants, Limiting Effec... Communication Remains A Fundamental Human Need, Yet Individuals With Hearing And Speech Impairments Face Persistent Barriers In Daily Interactions. Traditional Methods Like Sign Language Require Mutual Proficiency Between Participants, Limiting Effectiveness In Diverse Real- World Scenarios Where Hearing Individuals Lack Sign Language Knowledge. This Paper Proposes An Innovative, Affordable Wearable Communication Aid Designed As A Smart Glove To Bridge This Gap, Enabling Seamless Two-way Interaction For Disabled Users. The System Integrates Tilt Sensors Within A Comfortable Wearable Glove To Precisely Detect Finger Gestures And Hand Positions. These Gestures Map To A Library Of Predefined Messages (e.g., "Help," "Water," "Thank You") Processed By An ESP32 Microcontroller. Using User Datagram Protocol (UDP) For Low-latency Wireless Transmission, Detected Messages Instantly Appear As Text Or Speech Output On Connected Smartphones, Tablets, Or Dedicated Displays. This Facilitates Rapid, Reliable Communication With Hearing Individuals In Real-time Settings Such As Shopping, Medical Visits, Or Public Spaces. Complementing Gesture Input, The System Incorporates Speech-to-text Conversion, Capturing Nearby Spoken Language And Displaying It As Readable Text On The User's Device. This Bidirectional Functionality Supports Comprehensive Conversations, Breaking Traditional One-way Communication Limitations. The Design Prioritizes Portability (lightweight Glove Form-factor), Low Power Consumption For All-day Use, Intuitive Gesture Mapping For Quick Learning, And Real-time Responsiveness (<100ms Latency). By Combining Embedded Gesture Recognition With Modern Wireless Protocols, This Assistive Technology Enhances User Independence, Boosts Social Participation, And Reduces Isolation. Cost-effective Components Ensure Accessibility Across Socioeconomic Groups, While Modular Design Supports Future Enhancements Like AI Gesture Learning And Multi-language Support. This Solution Advances Wearable Assistive Technology, Promoting Inclusivity And Equal Communication Opportunities. Read more Read less

DEEP LEARNING – BASED LECTURE VIDEO SUMMARIZATION FOR SMART EDUCATION

The Rapid Growth Of Online Learning Platforms Has Led To A Large Collection Of Lecture Videos, Making It Difficult For Students To Revisit Lengthy Content. To Solve This Issue, The Proposed System Is Developed As A User-friendly Web Application That... The Rapid Growth Of Online Learning Platforms Has Led To A Large Collection Of Lecture Videos, Making It Difficult For Students To Revisit Lengthy Content. To Solve This Issue, The Proposed System Is Developed As A User-friendly Web Application That Converts Long Lecture Videos Into Short And Meaningful Summaries. The System Extracts Audio From Videos And Converts It Into Text Using Speech-to-text Technology, Which May Contain Redundant Information. Natural Language Processing (NLP) And Deep Learning Techniques Are Applied To Identify Key Concepts And Important Sentences. Based On This, A Structured Summary Is Generated To Highlight The Main Ideas Of The Lecture. This Helps Users Quickly Understand The Content Without Watching The Entire Video. The System Improves Learning Efficiency, Saves Time, And Supports Effective Revision. Overall, It Provides An Intelligent Solution For Managing And Summarizing Lecture Video Content. Read more Read less

NUTRIENT DEFICIENCY DETECTION IN PADDY CROP USING LEAF IMAGES

The Early Detection Of Nutrient Deficiencies In Paddy Crops Is Essential For Improving Crop Yield And Ensuring Sustainable Agricultural Practices. Traditional Methods Rely On Manual Observation And Expert Knowledge, Which Can Be Time-consuming, Costl... The Early Detection Of Nutrient Deficiencies In Paddy Crops Is Essential For Improving Crop Yield And Ensuring Sustainable Agricultural Practices. Traditional Methods Rely On Manual Observation And Expert Knowledge, Which Can Be Time-consuming, Costly, And Prone To Errors. This Paper Presents A Deep Learning Based Approach To Detect Nitrogen (N), Phosphorus (P), And Potassium (K) Deficiencies Using Paddy Leaf Images. The Proposed System Utilizes A MobileNetV2 Transfer Learning Model For Accurate Image Classification. Image Pre-processing Techniques Such As Resizing, Normalisation, And Data Augmentation Are Applied To Enhance Model Performance. The System Is Implemented As A Web-based Application Using HTML, CSS, JavaScript, And Python Flask, Allowing Users To Upload Leaf Images And Receive Real-time Predictions Along With Confidence Scores. Additionally, The System Provides Fertilizer Recommendations And Generates PDF Reports For Future Reference. The Model Achieves High Classification Accuracy, Demonstrating Its Effectiveness In Identifying Nutrient Deficiencies And Supporting Precision Agriculture. Read more Read less

AI-DRIVEN TWO WAY SIGN LANGUAGE TRANSLATOR

This Project Presents A Two-way Sign Language Translation System That Enables Communication Between Hearing-impaired Individuals And Non-sign Language Users. The System Converts Hand Gestures Into Text Using Computer Vision And Machine Learning Techn... This Project Presents A Two-way Sign Language Translation System That Enables Communication Between Hearing-impaired Individuals And Non-sign Language Users. The System Converts Hand Gestures Into Text Using Computer Vision And Machine Learning Techniques, And Also Translates Text Into Sign Language Through Visual Representation. The Sign-to-text Module Uses Real-time Hand Detection And Classification To Recognize Gestures, While The Text-to-sign Module Generates Corresponding Sign Outputs. The Model Is Trained Using Labeled Gesture Data And Optimized For Accuracy And Speed. This Solution Helps Bridge The Communication Gap And Supports Inclusive Interaction In Everyday Environments. Read more Read less

Detection Of Fake And Irrelevant Job Postings Using Passive Aggressive Classifier

With The Rapid Growth Of Online Job Portals, The Number Of Fake And Irrelevant Job Postings Has Significantly Increased. These Fraudulent Listings Mislead Job Seekers, Waste Time, And Sometimes Lead To Financial Loss. This Paper Presents A Machine Le... With The Rapid Growth Of Online Job Portals, The Number Of Fake And Irrelevant Job Postings Has Significantly Increased. These Fraudulent Listings Mislead Job Seekers, Waste Time, And Sometimes Lead To Financial Loss. This Paper Presents A Machine Learning–based Approach To Detect Fake And Irrelevant Job Postings Using A Passive Aggressive Classifier. A Dataset Of 10,000 Job Postings Collected From Kaggle Was Used For Training And Evaluation. Natural Language Processing (NLP) Techniques Such As TF-IDF Are Applied To Convert Textual Data Into Numerical Features. The Proposed System Is Integrated Into A Web Platform Named TrueHire, Which Provides Verified Job Listings. The Model Achieves An Accuracy Of 71.93%, Demonstrating Its Effectiveness In Identifying Fraudulent And Irrelevant Postings. Read more Read less

A COMPREHENSIVE REVIEW OF INTEGRATED SOLAR-WIND HYBRID RENEWABLE ENERGY SYSTEM FOR RELIABLE AND SUSTAINABLE RURAL POWER GENERATION

Rural Electrification Remains A Critical Global Challenge, With Approximately 759 Million People Lacking Access To Electricity, Primarily In Rural Regions Of Developing Countries. This Study Presents A Comprehensive Evaluation Of An Integrated Solar–... Rural Electrification Remains A Critical Global Challenge, With Approximately 759 Million People Lacking Access To Electricity, Primarily In Rural Regions Of Developing Countries. This Study Presents A Comprehensive Evaluation Of An Integrated Solar–wind Hybrid Renewable Energy System (HRES) As A Sustainable Solution For Rural Power Generation. The Proposed System Is Modeled And Optimized Using MATLAB/Simulink, Incorporating Photovoltaic (PV) Arrays, Wind Turbines, Battery Energy Storage, And Advanced Energy Management Controllers. Maximum Power Point Tracking (MPPT) Techniques Are Employed To Maximize Energy Extraction From Both Solar And Wind Sources, Utilizing The Perturb And Observe (P&O) Algorithm For PV Systems And The Tip Speed Ratio (TSR) Control Method For Wind Energy Conversion. Simulation Results Demonstrate That The Integrated System Provides Enhanced Reliability Compared To Standalone Renewable Sources Due To The Complementary Nature Of Solar And Wind Energy. Furthermore, The Hybrid Configuration Reduces Energy Storage Requirements By Up To 35%, Thereby Improving Overall System Efficiency And Cost-effectiveness. The Results Confirm The Technical Feasibility And Economic Viability Of Solar–wind Hybrid Systems For Sustainable Rural Electrification And Highlight Their Potential Contribution Toward Achieving The United Nations Sustainable Development Goals (SDGs) For Universal Energy Access. Read more Read less

AI-Based Automatic Network Traffic Routing To Avoid Congestion

The Exponential Increase In Network Traffic Has Made Congestion A Major Challenge, Leading To Increased Latency, Packet Loss, And Reduced Quality Of Service (QoS). This Paper Presents An Intelligent And Adaptive Traffic Routing System That Integrates... The Exponential Increase In Network Traffic Has Made Congestion A Major Challenge, Leading To Increased Latency, Packet Loss, And Reduced Quality Of Service (QoS). This Paper Presents An Intelligent And Adaptive Traffic Routing System That Integrates Software-Defined Networking (SDN), Python-based Automation, And Artificial Intelligence/Machine Learning (AI/ML) To Proactively Detect And Mitigate Network Congestion. The Proposed System Utilizes The Centralized Control Capability Of SDN To Continuously Monitor Network Conditions And Dynamically Manage Traffic Flows. Machine Learning Algorithms Are Trained Using Both Historical And Real-time Network Data To Accurately Predict Congestion Based On Key Performance Metrics Such As Bandwidth Utilization, Delay, And Packet Loss. Upon Identifying Potential Congestion, The System Automatically Selects Optimal Alternative Paths And Reroutes Traffic In Real Time, Ensuring Efficient Bandwidth Utilization And Reduced Network Delays. Experimental Evaluation Shows That The Integration Of AI/ML With SDN Significantly Improves Throughput, Reduces Congestion Levels, And Achieves Effective Load Balancing. This Research Demonstrates A Smart And Scalable Approach For Next-generation Networks, Suitable For Data Centers, Cloud Computing, And IoT Infrastructures. Read more Read less

Feasibility Study Of Solar Powered Water Pumping System

Solar-powered Water Pumping Systems Are An Efficient And Sustainable Alternative To Conventional Diesel And Electric Pumps. This Study Evaluates The Technical, Economic, And Environmental Feasibility Of Such Systems. Solar Photovoltaic (PV) Technolog... Solar-powered Water Pumping Systems Are An Efficient And Sustainable Alternative To Conventional Diesel And Electric Pumps. This Study Evaluates The Technical, Economic, And Environmental Feasibility Of Such Systems. Solar Photovoltaic (PV) Technology Converts Sunlight Into Electrical Energy To Operate Water Pumps, Making It Highly Suitable For Rural And Agricultural Applications. The Results Show That Although Initial Costs Are High, The System Offers Low Operating Cost, Long Lifespan, And Zero Emissions, Making It A Viable Long-term Solution. Read more Read less

IOT BASED PARALYSIS PATIENT HEALTH CARE USING ARDUINO UNO Wi-Fi MODULE

People Living With Complete Or Partial Paralysis Are Unable To Speak Or Generate Significant Body Movements, Leaving Them Entirely Reliant On The Constant Presence Of Attendants To Address Even The Most Basic Needs. Delays In Response — Whether Due T... People Living With Complete Or Partial Paralysis Are Unable To Speak Or Generate Significant Body Movements, Leaving Them Entirely Reliant On The Constant Presence Of Attendants To Address Even The Most Basic Needs. Delays In Response — Whether Due To Caregiver Unavailability Or Communication Barriers — Can Escalate Minor Discomforts Into Medical Emergencies. The Goal Of This Project Is To Develop A Low-cost, Embedded System-based Solution That Restores A Degree Of Communicative Autonomy To Such Patients By Converting Residual Hand Movements Into Intelligible Messages That Can Be Delivered Both Locally And Over The Internet. A Wearable Transmitter Unit Worn On The Patient's Wrist Houses An MPU-6050 Inertial Measurement Unit. When The Patient Tilts Or Rotates The Hand, The Sensor Captures Six-axis Motion Data That An Onboard ATmega328P Microcontroller Interprets Against A Library Of Pre-mapped Gesture Patterns. Each Recognised Pattern Triggers The Transmission Of A Coded Signal Over A 433 MHz Radio-frequency (RF) Link To A Corresponding Receiver Unit Placed At The Nursing Station Or Caregiver's Location. Healthcare Personnel And Family Members With Authorised Access Can Therefore Monitor The Patient's Requests And Alerts In Real Time From Any Internet-connected Device, Regardless Of Physical Proximity. Bench Testing Across 250 Gesture Trials Yielded An Average Recognition Accuracy Of 94.4 %, An RF Link Reliability Of 98.2 %, And An End-to-end Cloud Notification Latency Below One Second On A Standard Broadband Connection. The System Is Portable, Power-efficient, And Built Entirely From Commercially Available, Off-the-shelf Components, Making It Readily Replicable In Resource-constrained Clinical And Home-care Settings. Read more Read less

Extraction And Characterization Of Mango Seed Kernel Starch For The Development Of PH-Responsive Smart Bioplastic Films

The Escalating Accumulation Of Synthetic Petroleum-based Plastics Has Precipitated A Global Environmental Crisis, Driving The Search For Sustainable, Biodegradable, And Functional Alternatives. This Research Investigates The Extraction And Utilizatio... The Escalating Accumulation Of Synthetic Petroleum-based Plastics Has Precipitated A Global Environmental Crisis, Driving The Search For Sustainable, Biodegradable, And Functional Alternatives. This Research Investigates The Extraction And Utilization Of Starch From Mango Seed Kernel (MSK), A Prominent Agro-industrial Waste, As A Biopolymer Matrix For Intelligent Packaging. A PH-responsive Indicator Was Synthesized By Incorporating Anthocyanin-rich Extract From Brassica Oleracea (Red Cabbage) Into The MSK Starch Matrix. The Bioplastic Films Were Fabricated Via The Solution Casting Technique, Optimized With Glycerol As A Plasticizer And Citric Acid As A Cross-linking Agent To Enhance Mechanical Integrity And Reduce Water Sensitivity. Comprehensive Characterization, Including Fourier Transform Infrared Spectroscopy (FTIR) Performed At The South India Textile Research Association (SITRA), Coimbatore, Confirmed The Successful Formation Of Ester Bonds And The Homogeneous Distribution Of Components. The Films Exhibited Distinct Colorimetric Transitions From Pink (acidic) To Violet (neutral) And Green/yellow (alkaline), Correlating With PH Variations. Mechanical Testing Revealed A Tensile Strength Of 4.85 MPa And An Elongation At Break Of 38.5%, Suitable For Flexible Packaging Applications. The Integration Of Anthocyanins Provided Real-time Visual Sensing Capabilities, Suggesting These Films Are Promising Candidates For Smart Food Packaging Systems That Monitor Freshness While Mitigating Environmental Pollution Through The Valorization Of Agricultural Waste. Read more Read less

Assemble, Flying And Performance Evaluation Of Quadcopter (UAV) For Agriculture Application

Unmanned Aerial Vehicles (UAVs) Have Gained Significant Importance In Both Military And Civilian Applications Worldwide In Recent Years. However, Their Design Involves Complex Optimization Of Multiple Parameters And Engineering Decisions. This Paper... Unmanned Aerial Vehicles (UAVs) Have Gained Significant Importance In Both Military And Civilian Applications Worldwide In Recent Years. However, Their Design Involves Complex Optimization Of Multiple Parameters And Engineering Decisions. This Paper Presents A Comprehensive Approach To The Design, Manufacturing, And Performance Evaluation Of An Electrically Powered Radio-controlled Quadcopter Drone, Aimed At Achieving Stable Take-off, Efficient Cruise Flight, Safe Landing, And Maximum Payload Carrying Capacity. The Development Process Includes Several Stages: Conceptual Design, Structural Analysis, Performance Analysis, Material Selection, Fabrication, And Experimental Testing. The Quadcopter Was Designed With An Empty Weight Of 15.43 Lbs And A Maximum Take-off Weight Of 33.07 Lbs. The System Achieved A Maximum (flight) Height Of 220m, An Operational Range Of 2-4km, And A Lift/payload Capacity Of 40 Lbs. Additionally, The Drone Demonstrated A Cruising Speed Of 46 Ft/s And A Maximum Speed Of 78 Ft/s, With A Controlled Take-off Distance Equivalent To Standard Multirotor Vertical Lift Characteristics. Experimental Results Indicate That The Proposed Design Provides Stable Flight Performance, Efficient Payload Handling, And Reliable Operation Under Varying Conditions. This Study Contributes To The Development Of Cost-effective And High-performance UAV Systems Suitable For Real-world Applications Such As Surveillance, Agriculture, And Delivery. Read more Read less

EPILEPTIC SEIZURE RISK ANALYSIS USING EEG SIGNALS

Epileptic Seizure Is A Sudden Neurological Event Caused By Abnormal Electrical Activity In The Human Brain. These Events Can Significantly Disrupt Patient Safety And Quality Of Life. Early Prediction Of Such Events Is Essential For Enabling Timely Me... Epileptic Seizure Is A Sudden Neurological Event Caused By Abnormal Electrical Activity In The Human Brain. These Events Can Significantly Disrupt Patient Safety And Quality Of Life. Early Prediction Of Such Events Is Essential For Enabling Timely Medical Intervention And Continuous Monitoring. This Paper Proposes An Adaptive Attention-based Deep Learning Framework To Analyze Epileptic Seizure Risks Using Electroencephalogram (EEG) Signals. The Proposed Framework Incorporates A Statistical Signal Validation Module To Ensure That Only Reliable EEG Segments Are Considered For Further Processing. The System Employs An Attention-based Feature Extraction Mechanism That Enables The Model To Automatically Focus On Important Temporal Regions Of The EEG Signal. This Mechanism Helps In Capturing Meaningful Patterns Associated With Seizure Activity While Reducing The Influence Of Irrelevant Or Noisy Data. Unlike Conventional Approaches That Rely On Fixed Thresholds, The Proposed System Introduces An Adaptive Thresholding Mechanism That Dynamically Adjusts Decision Boundaries Based On Input Signal Characteristics, Improving Performance Across Different Patients. In Addition To Performing Seizure Detection And Prediction, The Proposed System Also Includes A Risk Level Classification Module That Categorizes EEG Signals Into Low, Medium, And High Levels Of Seizure Risk. Furthermore, A Visualization Component Is Integrated Into The System To Highlight Important Signal Regions That Influence The Model’s Decisions, Thereby Improving Interpretability. A Simple Decision Support Mechanism Is Also Incorporated To Enhance The Practical Usability Of The Prediction Results. The Experimental Results Demonstrate That The Proposed System Can Effectively Detect And Predict Seizures With Reliable Performance. The Framework Provides A Practical And Efficient Solution For Intelligent Seizure Risk Analysis Using EEG Signals. Read more Read less

Review Of Structural Performance: Pushover Vs Static Analysis Of Multistorey Buildings Using BIM

During Seismic Action, Building Will Deform In In-elastic Zone, So Its Required Evaluation Which Consider Post-elastic Behavior Of Structure. Performance Based Seismic Design Is A Modern Technique To Earthquake Resistance Which Can Predict Performanc... During Seismic Action, Building Will Deform In In-elastic Zone, So Its Required Evaluation Which Consider Post-elastic Behavior Of Structure. Performance Based Seismic Design Is A Modern Technique To Earthquake Resistance Which Can Predict Performance Of Structure Using Rigorous Non-linear Static Analysis. Easy And Most Used Method To Evaluate Performance Of Structure Is Non-linear Static Analysis Widely Known As Pushover Analysis. As Name Implies, It’s A Process Of Pushing Structure Horizontally, With A Prescribed Loading Pattern Incrementally, I.e. "pushing Structure & Plotting Total Applied Shear Force & Associated Lateral Displacement At Each Increment, Until Structure Reaches A Limit State Or Collapse Condition". This Project Aims To Analysis The Structural Behavior Of Multi-storey Building By Comparing The Result Of Static And Pushover Analysis Using Building Information Modeling (BIM) Implementation. BIM Is A Complex Process Of New Intelligent Approach And Process Of Maintaining All Relevant Information To A Building Over All Phase Of The Building Life Cycle. It Is Used To Improvement Of Process, Predict Outcomes And Create Computational Representation Of All Building With Less Environmental Impact. Read more Read less

Design And Development Of An Angular Drilling Fixture For Precision Machining Applications

This Research Paper Presents The Design, Development, And Implementation Of An Angular Drilling Fixture For Precision Machining Operations. Angular Drilling Fixtures Are Specialized Work-holding Devices That Enable Accurate Drilling At Predetermined... This Research Paper Presents The Design, Development, And Implementation Of An Angular Drilling Fixture For Precision Machining Operations. Angular Drilling Fixtures Are Specialized Work-holding Devices That Enable Accurate Drilling At Predetermined Angles On Workpieces, Addressing Significant Challenges In Conventional Drilling Operations Including Manual Alignment Errors, Inconsistent Hole Positioning, And Reduced Productivity. The Proposed Fixture Incorporates A Rigid Base Plate, Precision Locating Elements, Efficient Clamping Mechanisms, And An Angular Support Block Designed For A 30-degree Drilling Application. The Design Follows The 3-2-1 Locating Principle To Constrain All Six Degrees Of Freedom, Ensuring Deterministic Positioning And Repeatability. The Fixture Was Fabricated Using Mild Steel For The Base Plate And EN8 For Critical Components, With Hardened Drill Bushings For Tool Guidance. Experimental Validation Demonstrated Linear Accuracy Of ±0.1 Mm, Significant Reduction In Setup Time, And Consistent Hole Quality Across Multiple Workpieces. The Fixture Successfully Eliminates Manual Measurement Requirements, Reduces Operator Skill Dependency, And Enhances Production Efficiency In Mass Manufacturing Environments. Applications Include Automotive, Aerospace, And General Manufacturing Sectors Requiring Precise Angled Holes For Assembly And Functional Purposes. Read more Read less

Design Of Instrument Apparatus For Measurement Of Follower Displacement With Cam Rotation

This Research Paper Presents The Design, Development, And Analysis Of An Instrument Apparatus For Measuring Follower Displacement With Respect To Cam Rotation. Cam And Follower Mechanisms Are Fundamental Mechanical Systems Widely Employed In Engineer... This Research Paper Presents The Design, Development, And Analysis Of An Instrument Apparatus For Measuring Follower Displacement With Respect To Cam Rotation. Cam And Follower Mechanisms Are Fundamental Mechanical Systems Widely Employed In Engineering Applications To Convert Rotary Motion Into Reciprocating Or Oscillatory Motion With Precise Control. The Primary Objective Of This Study Is To Analyze The Kinematic And Dynamic Performance Of Cam-follower Systems, Focusing On Follower Displacement, Velocity, And Acceleration Characteristics For Different Cam Profiles. Various Motion Laws Including Uniform Velocity, Simple Harmonic Motion (SHM), And Cycloidal Motion Are Evaluated For Their Impact On Operational Smoothness, Jerk, And Vibration Levels. The Cam Profile Is Designed Using Both Graphical And Analytical Methods To Ensure Accurate Motion Transmission. The Study Also Examines Dynamic Aspects Including Contact Forces, Contact Stresses, Friction Effects, Lubrication Requirements, And Material Selection For Reliability And Durability. The Results Demonstrate That Appropriate Cam Profile Selection Significantly Reduces Shock, Minimizes Excessive Vibrations, And Improves Overall Efficiency. This Research Has Significant Applications In Internal Combustion Engines, Automated Machinery, Textile Machines, And Packaging Equipment Where Precise Timing And Motion Control Are Essential. Read more Read less

Speech-Driven Abstractive Summarization For Automated Meeting Minutes Generation

Meetings Are An Essential Part Of The Decision-making Process Of An Organization. However, Manually Writing The Minutes Of Meeting (MoM) Is A Laborious And Time-consuming Task. This Paper Proposes An Automated System For Generating The Minutes Of Mee... Meetings Are An Essential Part Of The Decision-making Process Of An Organization. However, Manually Writing The Minutes Of Meeting (MoM) Is A Laborious And Time-consuming Task. This Paper Proposes An Automated System For Generating The Minutes Of Meeting From The Meeting Audio. The System Uses Automatic Speech Recognition (ASR) And Natural Language Processing (NLP) For Generating A Well-structured And Concise Textual Summary Of The Meeting. The System Uses The Open AI Whisper ASR Model To Generate The Text From The Multi-speaker Meeting Audio. After Text Generation, The Text Is Processed, And Then The BERT-based Text Summarization Model Is Used To Generate A Well-structured MoM. The Results Of The Experiment Show That The Proposed System Can Generate The MoM Document Very Quickly, With Acceptable Accuracy. The System Can Be Used For Real-time And Offline Applications, Which Can Be Used For Academic And Corporate Environments. Read more Read less

Detection Of Screen Shadowing-Based Visual Data Exfiltration Attacks In VNC Systems

With The Increasing Adoption Of Remote Desktop Technologies In Enterprise Environments, The Security Of Visual Network Computing (VNC) Systems Has Become A Critical Concern. Screen Shadowing Attacks Represent A Significant Threat Vector Where Malicio... With The Increasing Adoption Of Remote Desktop Technologies In Enterprise Environments, The Security Of Visual Network Computing (VNC) Systems Has Become A Critical Concern. Screen Shadowing Attacks Represent A Significant Threat Vector Where Malicious Actors Silently Capture Sensitive Visual Data Displayed On Remote Desktops, Including Passwords, Financial Information, And Confidential Documents. This Paper Proposes The Design And Implementation Of A Real-time Detection System For Identifying Screen Shadowing-based Visual Data Exfiltration Attacks In VNC Environments. The Proposed System Incorporates A Virtual Laboratory Environment Consisting Of Ubuntu-based VNC Server And Kali Linux-based Attacker Systems Connected Through An Isolated Internal Network. Network Traffic Analysis Is Performed Using Wireshark And Tshark Tools To Establish Baseline Traffic Patterns During Normal VNC Usage. Attack Simulations Including Rapid Screen Capture And High-quality Stream Extraction Are Conducted To Generate Attack Traffic Signatures. A Python-based Detection Engine Utilizing Threshold-based Anomaly Detection And Machine Learning Algorithms, Specifically Isolation Forest, Is Implemented To Identify Deviations From Normal Traffic Patterns. The System Provides Real-time Alerting Capabilities And Automated Evidence Capture For Forensic Analysis. Experimental Results Demonstrate That The Proposed System Achieves High Detection Accuracy With Minimal False Positives, Effectively Identifying Screen Shadowing Attacks Through Bandwidth Analysis, Packet Rate Monitoring, And Statistical Pattern Recognition. The Proposed Architecture Provides Organizations With An Effective Tool For Protecting Sensitive Visual Data In VNC-enabled Remote Work Environments. Read more Read less

AI-Based Forensic Face Sketch Generation and Suspect Identification from Eyewitness Verbal Description

Suspect Identification Based On Eyewitness Descriptions Remains A Critical Challenge In Criminal Investigations Due To Memory Limitations, Language Barriers, And The Lack Of Reliable Visual Evidence. Traditional Forensic Sketching Methods Rely On Ski... Suspect Identification Based On Eyewitness Descriptions Remains A Critical Challenge In Criminal Investigations Due To Memory Limitations, Language Barriers, And The Lack Of Reliable Visual Evidence. Traditional Forensic Sketching Methods Rely On Skilled Artists, Making The Process Time-consuming, Subjective, And Often Inconsistent. To Address These Challenges, This Paper Presents An AI-Based Forensic Face Sketch Generation And Suspect Identification System For Automated And Efficient Investigation Support. The Proposed Framework Integrates A Multilingual Speech-to-text Module Using Whisper To Convert Eyewitness Descriptions Into Text. These Features Are Processed Using An Attention-Based Conditional Generative Adversarial Network (Attention-cGAN) To Generate Realistic Forensic Face Sketches, Which Are Further Enhanced Into Photo-like Images. The Generated Faces Are Encoded Using A Vision Transformer (ViT) To Extract Feature Embeddings And Matched With Criminal Database Records Using Cosine Similarity. The System Supports Multilingual Input And Provides Automated Result Visualization With Similarity Scores. Experimental Results Demonstrate Effective Performance In Generating Identity-consistent Facial Representations And Improving Matching Accuracy. The Integration Of Speech Processing, Generative Modeling, And Feature-based Matching Makes The System Suitable For Real-world Forensic Applications. Read more Read less

Study Of Deck Bridge Culverts Under Different Mix Proportions And Span Configurations

Culverts Are Required To Be Provided Under Earth Embankment For Crossing Of Water Course Like Streams, Nallas Etc. Across The Embankment, As Road Embankment Cannot Be Allowed To Obstruct The Natural Water Way. This Paper Deals With The Study Of Desig... Culverts Are Required To Be Provided Under Earth Embankment For Crossing Of Water Course Like Streams, Nallas Etc. Across The Embankment, As Road Embankment Cannot Be Allowed To Obstruct The Natural Water Way. This Paper Deals With The Study Of Design Parameters Of Box Culverts Like Effect Of Co-efficient Of Earth Pressure, Angle Of Dispersion Of Live Load And Depth Of Cushion Provided On Top Slab Of Box Culverts. Present Study Has Been Performed To Know How Design Of IRC-112 Differs From IRC-21 And An Attempt Is Made To Study Undefined Parameters Of IRC: 112- 2011 Such As Span To Depth (L/d) Ratio. Present Study Is Performed On Design Of RC Slab Culvert Using “working Stress Method” Using “IRC: 21-2000 And Limit State Method Using IRC: 112-2011” Code Specifications. It Is Observed That In Working Stress Method, The Allowable L/d Ratio Is 13 And In Limit State Method, The L/d Ratio Of 20 Is Most Preferable. This Paper Deals With The Study Of Design Parameters Of Box Culverts Like Effect Of Co-efficient Of Earth Pressure, Angle Of Dispersion Of Live Load And Depth Of Cushion Provided On Top Slab Of Box Culverts. This Paper Presents An Overview Of The Design Construction, And Laboratory And Field Testing Of A Box Culvert Bridge Reinforced With Glass FRP (GFRP) Bars. Here An Effort Has Been Made To Shows The Economic And Effective Design Can Be Achieved By Doing Finite Element Analysis Element Analysis Of A Box Culvert Whose Concept Can Be Used For Large Structural Design As Well. Read more Read less

Evaluation And Design Of Foundations Under Earthquake Forces As Per Indian And Euro Codes

The Present Study Focuses On The Analytical Evaluation And Design Of Footings Subjected To Seismic Forces Using Both Indian Standard And Eurocode Provisions With The Aid Of STAAD.Pro Software. A G+10 Storey Building Is Modeled In STAAD.Pro V8i, And P... The Present Study Focuses On The Analytical Evaluation And Design Of Footings Subjected To Seismic Forces Using Both Indian Standard And Eurocode Provisions With The Aid Of STAAD.Pro Software. A G+10 Storey Building Is Modeled In STAAD.Pro V8i, And Pad Footings Are Designed Considering Two Soil Conditions, Namely Hard And Medium Soils. Further, A Comparative Assessment Between Indian And Euro Standards Is Carried Out Using STAAD Foundation Software To Understand The Influence Of Different Code Provisions On Foundation Behavior Under Seismic Loading.The Analysis Aims To Examine How Various Foundation Systems Respond To Earthquake Forces And To Identify Efficient And Economical Design Approaches. Understanding The Effect Of Soil Conditions And Code-based Design Differences Plays A Crucial Role In Ensuring Structural Safety And Performance.Structural Design And Analysis Are Largely Influenced By Geographical Conditions, As Natural Hazards Vary From Region To Region. Among These, Seismic Forces Are One Of The Most Critical Factors Responsible For Structural Damage, Leading To Significant Economic Losses And Threats To Human Life. Therefore, Engineers Must Possess A Strong Understanding Of Different Design Codes And Apply Them Effectively To Develop Safe, Economical, And Long-span Structures.In Developing Countries, Construction Plays A Vital Role In Infrastructure Growth. Each Country Adopts Its Own Set Of Design Codes That Guide Engineers In Designing Key Structural Components Such As Beams, Columns, Slabs, And Foundations, Ensuring Safety, Reliability, And Standardization In Construction Practices. Read more Read less

STUDY OF PERMEABLE BRICK PAVEMENT

Permeable Brick Pavement (PBP) Is An Innovative And Sustainable Pavement System Designed To Allow Water Infiltration Through Its Surface Into Underlying Layers. Rapid Urbanization Has Increased Impervious Surfaces, Leading To Flooding, Groundwater De... Permeable Brick Pavement (PBP) Is An Innovative And Sustainable Pavement System Designed To Allow Water Infiltration Through Its Surface Into Underlying Layers. Rapid Urbanization Has Increased Impervious Surfaces, Leading To Flooding, Groundwater Depletion, And Environmental Degradation. This Research Paper Studies The Structure, Working Mechanism, Materials, Design Considerations, Advantages, Limitations, And Applications Of Permeable Brick Pavements. The Study Highlights That PBP Significantly Reduces Stormwater Runoff, Improves Groundwater Recharge, And Enhances Environmental Sustainability. However, Challenges Such As Clogging, Maintenance, And Structural Limitations Restrict Its Widespread Application. The Paper Concludes That Permeable Brick Pavements Are An Effective Solution For Modern Sustainable Urban Infrastructure When Properly Designed And Maintained. Read more Read less

AI-Synthesised Solo Podcast Platform For Knowledge Enhancement

Digital Content Creation Has Seen A Massive Surge, Yet Solo Creators Often Face Technical Hurdles In Scriptwriting And Professional Audio Production. This Paper Presents 'Podcastr', An Innovative Platform Designed To Automate The End-to-end Podcastin... Digital Content Creation Has Seen A Massive Surge, Yet Solo Creators Often Face Technical Hurdles In Scriptwriting And Professional Audio Production. This Paper Presents 'Podcastr', An Innovative Platform Designed To Automate The End-to-end Podcasting Lifecycle. Unlike Traditional Systems That Rely On Manual Recording, Our Proposed System Integrates The Groq LPU (Language Processing Unit) For Near-instant Script Generation Using The Llama 3 Model. For Audio Synthesis, We Utilize The Microsoft Edge TTS Engine To Deliver Human-like Neural Voices, Eliminating The Need For Expensive Studio Equipment. The Backend Is Architected Using FastAPI To Ensure Asynchronous Processing Of AI Tasks, While Hotwire Turbo Is Implemented To Maintain Persistent Audio Playback During Site Navigation. Our Results Demonstrate A Significant Reduction In Production Time—from Hours To Mere Seconds—making Professional Podcasting Accessible To Individual Knowledge-sharers. Read more Read less

Learning Path Dashboard For Enhancing Skills: A Personalized Adaptive E-Learning System

Personalized Learning Is Increasingly Important In Modern Education, Addressing Diverse Learning Styles And Indi-vidual Skill Gaps. This Project Presents A Frontend Prototype Of A Learning Path Dashboard That Simulates AI-inspired Adaptive Learning U... Personalized Learning Is Increasingly Important In Modern Education, Addressing Diverse Learning Styles And Indi-vidual Skill Gaps. This Project Presents A Frontend Prototype Of A Learning Path Dashboard That Simulates AI-inspired Adaptive Learning Using Mock Data And Rule-based Logic. The System Demonstrates Visualization Of Learning Trajectories, Skill Progres-sion Tracking, Interactive Learning Path Generation, And Mentor Guidance Without Requiring Backend Infrastructure Or AI Imple-mentation. The Prototype Validates UI Flow, Feature Integration, And User Experience Design, Providing A Foundation For Future AI-driven Recommendation Systems And Backend Services. Read more Read less

AI POWERED PRODUCTIVITY PARTNER

The Advancement Of Artificial Intelligence Has Led To The Emergence Of Autonomous Agents, Which Are Intelligent Systems Developed To Improve User Productivity, Decision-making, And Task Management. These Systems Combine Techniques Such As Natural Lan... The Advancement Of Artificial Intelligence Has Led To The Emergence Of Autonomous Agents, Which Are Intelligent Systems Developed To Improve User Productivity, Decision-making, And Task Management. These Systems Combine Techniques Such As Natural Language Processing, Machine Learning, And Context-aware Automation To Support Users In Activities Including Scheduling, Data Retrieval, Prioritization, And Performance Monitoring. By Learning From User Interactions And Adapting To Individual Preferences, Autonomous Agents Provide Personalized Suggestions, Automate Repetitive Operations, And Minimize Mental Effort. Read more Read less

Seismic Performance Evaluation Of Industrial Steel Structures Through Dynamic Analysis

The Study Concludes That Incorporating Bracing Systems And Dampers Significantly Enhances The Seismic Performance Of Industrial Steel Structures By Effectively Reducing Lateral Displacement, Vibration, And Storey Drift. Among The Configurations Analy... The Study Concludes That Incorporating Bracing Systems And Dampers Significantly Enhances The Seismic Performance Of Industrial Steel Structures By Effectively Reducing Lateral Displacement, Vibration, And Storey Drift. Among The Configurations Analyzed, Models With X-bracing Combined With Dampers Having A Mass Ratio Of 1.5% Demonstrated Superior Performance Under Both Earthquake And Wind Loading Conditions. A Comparative Assessment Between Conventional Steel Frames, Braced Systems, And Damper-integrated Structures Highlights The Effectiveness Of These Lateral Load-resisting Strategies. This Research Involves A Detailed Parametric Study Based On Nonlinear Dynamic Analysis Of Three-dimensional Industrial Steel Structures Using SAP2000 Software. Various Bracing Configurations, Particularly X-bracing Systems, Along With Different Damper Mass Ratios, Are Considered To Evaluate Their Influence On Structural Response Under Seismic Excitation. Industrial Steel Structures Often Possess Large Member Sizes, Resulting In Higher Dead Loads, Which In Turn Increase Their Susceptibility To Seismic Effects. Despite The Growing Adoption Of Braced Systems, Comprehensive Analytical Studies Addressing Their Effectiveness In Controlling Seismic Response Remain Limited In India. The Primary Objective Of This Work Is To Identify And Propose Efficient And Practical Lateral Load-resisting Systems (LLRSS) Suitable For Industrial Buildings. These Systems Aim To Improve Structural Safety And Can Be Applied To Both New Constructions And Existing Structures. Since Buildings Are Subjected To Lateral Loads From Earthquakes And Wind In Addition To Gravity Loads, Inadequate Design Against Such Forces May Lead To Excessive Stresses, Structural Sway, And Potential Collapse, Posing Risks To Life And Property. Therefore, It Is Essential To Adopt Appropriate Structural Technologies That Ensure Stability And Resilience Under Lateral Loading Conditions. Read more Read less

Safefaceyolo: Real-time Workplace Safety Enforcement Via Integrated Helmet Detection And Facial Recognition

Workplace Safety Is A Critical Concern In Industries Such As Construction And Manufacturing, Where Workers Are Exposed To Hazardous Environments. One Of The Major Causes Of Accidents Is The Failure To Wear Personal Protective Equipment (PPE), Particu... Workplace Safety Is A Critical Concern In Industries Such As Construction And Manufacturing, Where Workers Are Exposed To Hazardous Environments. One Of The Major Causes Of Accidents Is The Failure To Wear Personal Protective Equipment (PPE), Particularly Safety Helmets. Traditional Safety Monitoring Systems Rely Heavily On Manual Inspection, Which Is Time-consuming, Costly, And Prone To Human Error, Especially In Large-scale Sites With Multiple Ongoing Activities. To Address These Challenges, This Project Proposes An Automated System That Integrates Computer Vision Techniques For Real-time Helmet Detection And Worker Identification. The System Combines Two Key Functionalities: Helmet Detection And Face Recognition. Helmet Detection Is Performed Using Advanced Object Detection Algorithms To Identify Whether Workers Are Wearing Safety Helmets, While Facial Recognition Is Used To Identify Individuals By Comparing Extracted Facial Features With A Pre-existing Database. The Proposed Approach Improves Efficiency By Eliminating The Limitations Of Manual Monitoring And Enabling Continuous Surveillance. It Also Ensures Better Safety Compliance By Providing Accurate And Real-time Detection Under Varying Environmental Conditions. The System Is Further Evaluated Based On Performance Metrics Such As Accuracy And Recall Rate Across Different Visual Scenarios Found On Construction Sites. Overall, This Integrated Solution Offers A Reliable And Intelligent Approach To Enhance Workplace Safety, Reduce Accidents, And Support Safety Inspectors Through Automated Monitoring And Immediate Detection Of Violations. Read more Read less

Assessment Of Fire-Induced Structural Behavior In A Single Bay Frame Using Multiple Codes

Fire Hazards Pose A Significant Threat To Structures In India, Often Leading To Severe Damage And, In Some Cases, Structural Collapse Due To Elevated Temperatures. This Study Primarily Focuses On The Behavior Of Structural Members, Particularly Beams... Fire Hazards Pose A Significant Threat To Structures In India, Often Leading To Severe Damage And, In Some Cases, Structural Collapse Due To Elevated Temperatures. This Study Primarily Focuses On The Behavior Of Structural Members, Particularly Beams And Columns, Under Fire Conditions. While The Indian Standard (IS) Codes Provide A Standard Fire Curve For Heavy Steel Reinforcement, There Is Limited Guidance Available For Reinforced Concrete Members. To Address This Gap, Multiple International Codes Are Considered, Including IS Code, Eurocode, The National Building Code Of Canada, And The National Construction Code Of Australia. Based On These Codes, Structural Models Are Developed In SAP Software Using Relevant Parameters, With Uniform Temperature Conditions Applied Throughout The Structure. The Models Are Analyzed Iteratively To Assess Failure Conditions And To Refine The Design For Improved Safety. Subsequently, Selected Beam And Column Sections Are Subjected To Detailed Thermal Analysis Using SAFIR Software, Which Operates As A Module Within The GiD Environment. The Thermal Response Results Are Further Processed Using DIAMOND Software To Obtain Detailed Insights. Finally, A Comparative Evaluation Of The Single-bay Frame Under Different Code Provisions Is Presented Through Graphical Representations. The Study Highlights The Differences In Structural Response Based On Various Fire Curves And Design Standards. Read more Read less

Pushover Analysis Approach For Evaluating And Designing Water Tanks With Diverse Staging Arrangements: A Review

The Present Study Investigates The Formulation Of Key Factors For The Modifying Factor Of The Seismic Response Of The RCC Framed Staging Of The Elevated Water Reservoir. The Analysis Revealed That Three Main Factors, Called Reserved Force, Ductility... The Present Study Investigates The Formulation Of Key Factors For The Modifying Factor Of The Seismic Response Of The RCC Framed Staging Of The Elevated Water Reservoir. The Analysis Revealed That Three Main Factors, Called Reserved Force, Ductility And Redundancy, Affect The Real Value Of The Response Change Factor And Therefore Must Be Taken Into Account When Determining The Change Of The Appropriate Response To Be Used During The Seismic Design Process. The Evaluation Of The Response Modification Factor Is Carried Out Using A Non- Linear Static Pushover Analysis. Pushover Analysis Is An Advanced Tool For Performing Static Nonlinear Analyzes Of Framed Structures. It Is Used To Evaluate Non-linear Behavior And Provides The Sequence And Mechanism For Forming The Plastic Hinge. Here, Displacement-controlled Displacement Analysis Is Used To Apply Seismic Forces In C.G. Of Container. The Thrust Curve, Which Is A Basic Cut Graph Relative To The Displacement Of The Roof, Provides The Effective Capacity Of The Structure In The Non-linear Range. This Could Be Due To The Lack Of Knowledge About The Correct Behavior Of The Tank Support System Due To The Dynamic Effect And Also Due To The Inadequate Geometric Selection Of The Staging. The Objective Of This Study Is To Understand The Behavior Of Different Stages, Under Different Load Conditions And To Reinforce The Conventional Type Of Staging, In Order To Obtain Better Performance During An Earthquake. This Paper Presents A Seismic Analysis Of Elevated Water Tanks Supported In Different Staging Models With Different Tank Storage Capacities. Here We Compare Two Different Support Systems, Such As Radial Reinforcement And Transverse Reinforcement, With The Basic Support System For Various Fluid Level Conditions. Eleven Models Are Used To Calculate The Shearing Of The Base And The Nodal Displacements For Staging With STAAD Pro. After Calculating The Base Cut And The Nodal Displacements Of Eleven Models For Empty And Complete Conditions. Three Different Types Of Staging Systems Were Analyzed. Read more Read less

Integrated Study On Loading And Structural Behavior Of Electric Pylon Structures With Height Variation

Electric Pylon Is A Structure Made Of Steel Having Relatively High Ratio For Its Length And Width. The Structure Is Strong Enough To Withstand Gravity And Several Type Of External Forces. Electric Pylon Is One Of The Vital Components Of Electric Tran... Electric Pylon Is A Structure Made Of Steel Having Relatively High Ratio For Its Length And Width. The Structure Is Strong Enough To Withstand Gravity And Several Type Of External Forces. Electric Pylon Is One Of The Vital Components Of Electric Transmission Line That Make A Transmission Tower. The Tower Must Be Reliable And Able To Operate Under Several Loading Conditions. Environmental Loading Contributes The Largest Percentage Of The Total Loads Experienced By Any Typical Tower Structure. This Paper Investigates The Loading Condition Usually Experiences By Typical Pylon Structure. The Structure Modeling Is Done Using Finite Element Software. The Influence Of Various Loading Conditions On The Structure Is Investigated And The Critical Elements Of The Tower Structure Are Determined. From The Result Obtained, It Shows That The Most Critical Element Is Bracing Member Situated On The Lowest Level Of The Pylon Structure. The Result Of This Study Also Shows That The Safety Level In The Entire Structure Is Acceptable Under All The Most Severe Loading Conditions. Read more Read less

SECURE CERTIFICATE VERIFICATION VIA HYBRID AES-ECC ENCRYPTION, FACE BIOMETRICS & BLOCKCHAIN

Digital Certificate Management Faces Persistent Threats Including Forgery, Unauthorised Access, And Data Manipulation. Conventional Centralised Systems Are Vulnerable To Single-point Failures, Poor Scalability, And Weak Identity Verification. This Pa... Digital Certificate Management Faces Persistent Threats Including Forgery, Unauthorised Access, And Data Manipulation. Conventional Centralised Systems Are Vulnerable To Single-point Failures, Poor Scalability, And Weak Identity Verification. This Paper Proposes A Secure Certificate Verification Framework Integrating Three Complementary Layers: Facial Biometric Authentication Using The Grassmann Manifold Algorithm, Hybrid Encryption Combining AES-256 For Bulk File Protection And Elliptic Curve Cryptography (ECC) For Secure Key Management, And A Blockchain-based Decentralised Immutable Ledger For Certificate Storage. One-Time Password (OTP) Delivery And SHA-256 Hash Validation Reinforce Access Control At Every Stage. A Flask-Python Web Prototype Demonstrates End-to-end Certificate Issuance, Encrypted Storage, Biometric-gated Retrieval, And Third-party Verification. The System Eliminates Dependence On Centralised Infrastructure, Provides Tamper-proof Transparency, And Scales Efficiently Across Multiple Institutions. Read more Read less

Seismic Efficiency Of Multistorey Structures: Reviewing Floating Column And Conventional Column Systems

The Seismic Performance Of Multistorey Buildings Is A Critical Concern In Earthquake- Prone Areas, As Structural Design Directly Impacts Resilience And Safety. Among Various Structural Configurations, Floating Columns (FCs) And Regular Columns (RCs)... The Seismic Performance Of Multistorey Buildings Is A Critical Concern In Earthquake- Prone Areas, As Structural Design Directly Impacts Resilience And Safety. Among Various Structural Configurations, Floating Columns (FCs) And Regular Columns (RCs) Present Distinct Behaviors Under Seismic Loads. This Paper Provides A Comprehensive Review Of Studies Analyzing The Seismic Response Of Buildings With FCs Compared To Those With RCs. Floating Columns, Often Used In Architectural Designs For Space Efficiency And Aesthetic Appeal, Can Significantly Influence A Building’s Lateral Strength And Stability Due To The Transfer Of Loads Onto Beams Rather Than Directly To The Foundation. This Review Focuses On Key Parameters Such As Lateral Displacement, Base Shear, Inter-storey Drift, And Natural Period, Comparing Their Impact On Overall Building Stability In Both FC And RC Configurations. Results From Various Analytical And Simulation-based Studies Reveal That Buildings With FCs Exhibit Increased Lateral Displacement And Inter- Storey Drift, Resulting In Higher Vulnerability To Seismic Events. In Contrast, Buildings With RCs Generally Demonstrate Better Seismic Performance And Structural Integrity. This Comparative Analysis Provides Insights Into Optimizing Structural Design To Improve Seismic Resilience, Offering Guidance For Engineers And Architects In Designing Safer, More Robust Multistorey Structures. Read more Read less

AN INTERPRETABLE FINANCIAL DECISION SUPPORT SYSTEM FOR LOAN APPROVAL USING DATA ANALYTICS

The Process Of Loan Approval Plays A Vital Role In Financial Institutions, Requiring Accurate And Timely Decision-making. Traditional Systems Rely Heavily On Manual Evaluation, Which Often Leads To Inconsistency And Delays. This Study Presents An Int... The Process Of Loan Approval Plays A Vital Role In Financial Institutions, Requiring Accurate And Timely Decision-making. Traditional Systems Rely Heavily On Manual Evaluation, Which Often Leads To Inconsistency And Delays. This Study Presents An Interpretable Financial Decision Support System For Loan Approval Using Data Analytics Techniques. The System Leverages Historical Applicant Data To Predict Loan Eligibility With Improved Precision. Key Attributes Such As Income, Credit Score, Loan Amount, And Asset Values Are Considered During Analysis. Machine Learning Models Are Applied To Identify Patterns And Relationships Within The Data. Among Various Algorithms, Random Forest Is Emphasized For Its Interpretability And Performance. The Proposed System Minimizes Human Bias And Enhances Transparency In Decisions. A User-friendly Interface Enables Real-time Prediction Of Loan Approval Status. The Overall Approach Improves Efficiency, Consistency, And Reliability In Financial Decision-making. The Model Is Designed To Be Scalable And Adaptable To Different Financial Datasets. It Also Supports Data-driven Policy Formulation In Modern Banking Systems.: Read more Read less

AI Powered Virtual Job Interview Simulator

The Increasing Demand For Effective Interview Preparation Tools Highlights The Limitations Of Traditional Methods, Which Often Lack Personalization And Real-time Feedback. This Paper Presents An AI-powered Virtual Job Interview Simulator Designed To... The Increasing Demand For Effective Interview Preparation Tools Highlights The Limitations Of Traditional Methods, Which Often Lack Personalization And Real-time Feedback. This Paper Presents An AI-powered Virtual Job Interview Simulator Designed To Provide A Realistic And Adaptive Interview Experience. The System Analyzes User Resumes To Extract Relevant Skills And Generates Context-aware Interview Questions Using Natural Language Processing Techniques, Ensuring That Interview Sessions Are Tailored To Individual Profiles. User Responses Are Evaluated In Real Time Based On Multiple Parameters, Including Grammatical Correctness, Semantic Relevance, Fluency, And Communication Clarity, Enabling A Comprehensive Assessment Of Candidate Performance. A Computer Vision-based Module Is Integrated For Face Verification And User Monitoring, Where A Hybrid Approach Improves Authentication Reliability. The System Also Provides Structured Feedback And Performance Analytics To Support Continuous Improvement Over Repeated Sessions. Experimental Results Demonstrate Reliable Performance In Question Relevance, Response Evaluation Consistency, And Face Verification Accuracy, While Maintaining Low Response Time. Overall, The Proposed System Offers An Efficient And Scalable Platform For Enhancing Interview Readiness, Improving Communication Skills, And Boosting User Confidence. Read more Read less

AURA-Adaptive User Response Assistant For Workplace Well-being

In Modern Digital Work Environments, Employee Stress Has Become A Major Factor Affecting Productivity And Organizational Performance. This Paper Presents AURA (Adaptive User Response Assistant), An Intelligent System Designed To Monitor Employee Stre... In Modern Digital Work Environments, Employee Stress Has Become A Major Factor Affecting Productivity And Organizational Performance. This Paper Presents AURA (Adaptive User Response Assistant), An Intelligent System Designed To Monitor Employee Stress Levels In Real Time Using Artificial Intelligence (AI), Machine Learning (ML), And Computer Vision (CV). The System Analyses Facial Expressions, Eye Movements, And Head Posture Using Webcam Input To Detect Stress Indicators. It Classifies Stress Levels Into Low, Medium, And High Categories And Generates Real-time Alerts To Encourage Employees To Take Breaks. The System Also Stores Stress Patterns In A Database And Provides Insights Through An HR Dashboard. The Proposed System Improves Workplace Well-being, Reduces Burnout, And Enhances Productivity Through Automated Monitoring And Data-driven Decision-making. Read more Read less

Secure Health Risk And Appointment System

Efficient Management Of Healthcare Services Remains A Significant Challenge, Particularly In Environments Where Coordination Between Patients, Hospital Branches, And Emergency Services Is Handled Through Manual Or Disconnected Systems. Issues Such As... Efficient Management Of Healthcare Services Remains A Significant Challenge, Particularly In Environments Where Coordination Between Patients, Hospital Branches, And Emergency Services Is Handled Through Manual Or Disconnected Systems. Issues Such As Appointment Delays, Lack Of Hospital Availability Visibility, And Inefficient Ambulance Coordination Directly Impact Patient Care. This Paper Presents A Web-based Secure Health Risk & Appointment System Providing A Centralized Platform For Managing Healthcare Operations. The System Enables Patients To Book Appointments Through A Structured Interface, While Hospital Administrators Monitor Doctor Availability And Update Operational Status Across Multiple Branches. Ambulance Staff Receive A Dedicated Dashboard For Emergency Alerts, Location Updates, And Hospital Selection Based On Availability. A Map-based Interface Visualizes Hospital And Ambulance Positions For Improved Emergency Coordination. Essential Cybersecurity Measures Including Secure Authentication, Role-based Access Control, Input Validation, And Activity Logging Ensure System Integrity. The Proposed System Demonstrates Improved Coordination, Reduced Delays, And Better Transparency Compared To Traditional Manual Approaches, Making It A Practical And Scalable Solution For Modern Healthcare Environments. Read more Read less

Deep Learning-Enabled Personal Fitness Coaching With Motion Feedback And Goal Optimization

The Increasing Adoption Of Artificial Intelligence In Healthcare And Fitness Has Opened New Opportunities For Intelligent Personal Coaching Systems. This Paper Presents A Deep Learning-enabled Personal Fitness Coaching System That Recognizes Exercise... The Increasing Adoption Of Artificial Intelligence In Healthcare And Fitness Has Opened New Opportunities For Intelligent Personal Coaching Systems. This Paper Presents A Deep Learning-enabled Personal Fitness Coaching System That Recognizes Exercise Activities And Provides Personalized Workout Recommendations. The Proposed System Uses A Custom-trained YOLOv11 Convolutional Neural Network (CNN) Model To Detect And Classify 36 Different Exercise Postures From User-uploaded Images And Videos. Based On The Detected Activity And User-provided Age, The System Dynamically Generates Personalized Workout Plans And Posture Correction Guidance. The Application Includes Modules Such As User Login, Deep Learning Model Training, Training Graph Visualization, Activity Recognition, And Plan Recommendation. Experimental Results Show That The Proposed System Provides Accurate Activity Recognition And Helps Users Improve Exercise Performance, Reduce Injury Risk, And Achieve Better Fitness Outcome. Read more Read less

Optimization Of Secure And Efficient Encrypt Image Retrieval Based On Additive Sharing Using SMPC

Due To Growing Threats Of Data Interception And Illegal Access, Secure Digital Image Transmission Over Open Networks Continues To Be A Major Concern. By Combining Elliptic Curve Cryptography (ECC), Visual Cryptography, And Least Significant Bit (LSB)... Due To Growing Threats Of Data Interception And Illegal Access, Secure Digital Image Transmission Over Open Networks Continues To Be A Major Concern. By Combining Elliptic Curve Cryptography (ECC), Visual Cryptography, And Least Significant Bit (LSB)-based Steganography, This Article Offers A Strong Framework For Safe Multi-image Transmission. Strong Encryption With A Smaller Key Size And Less Computational Cost Is Provided By ECC, Guaranteeing Effective And Safe Key Management. Visual Cryptography Is Used To Further Process The Encrypted Images, Breaking Each One Up Into Several Pieces That Don't Independently Give Any Useful Information. LSB Steganography Is Used To Embed These Files Inside Cover Images In Order To Improve Confidentiality. This Effectively Hides The Existence Of Sensitive Data During Transmission. The Original Images Are Recovered With High Fidelity At The Receiving End By Extracting The Embedded Shares, Reconstructing Them Using Visual Cryptography, And Decrypting Them Using ECC. Improved Peak Signal-toNoise Ratio (PSNR) Data Show That The Suggested Method Achieves High Reconstruction Accuracy, Minimum Visual Distortion, And Superior Security. Strong Defense Against Statistical Analysis, Cryptographic Attacks, And Illegal Access Is Offered By The Multi-layered Security System. All Things Considered, The Suggested System Provides A Scalable, Effective, And Extremely Secure Multi-image Transmission Solution, Making It Appropriate For Use In Defense Systems, Medical Imaging, And Secure Communications. Read more Read less

REAL-TIME VIOLENCE DETECTION SYSTEM USING YOLOV11 AND DEEP LEARNING

The Widespread Deployment Of Surveillance Systems Has Resulted In Continuous Generation Of Large-scale Video Data, Creating Challenges In Timely Identification Of Violent Incidents Such As Physical Assaults And Aggressive Behavior. Conventional Monit... The Widespread Deployment Of Surveillance Systems Has Resulted In Continuous Generation Of Large-scale Video Data, Creating Challenges In Timely Identification Of Violent Incidents Such As Physical Assaults And Aggressive Behavior. Conventional Monitoring Approaches That Rely On Manual Observation Or Rule-based Methods Often Suffer From Delays, Inaccuracies, And High Dependency On Human Intervention. To Address These Limitations, This Research Presents A Real-time Violence Detection Framework Based On Advanced Deep Learning Techniques. The Proposed Approach Employs The YOLOv11 Algorithm For Rapid Object Detection, Enabling Efficient Identification Of Weapons And Suspicious Activities In Video Streams. Extracted Frames Are Processed Through Convolutional Neural Networks To Capture Contextual And Behavioral Features Associated With Violent Actions. The Integration Of Spatial And Temporal Analysis Enhances The System’s Capability To Differentiate Between Normal And Abnormal Human Behavior, Thereby Reducing False Alarms. Upon Detection Of Potential Threats, The System Generates Real-time Alerts Containing Essential Information Such As Timestamps, Detected Objects, And Confidence Levels. This Facilitates Prompt Response From Security Personnel And Improves Overall Situational Awareness. The Research Demonstrates An Effective And Scalable Solution For Automated Surveillance, Contributing To Enhanced Public Safety And Reliable Monitoring In Complex Environments. Read more Read less

Integrated GAN And IR Algorithms For Noise-Robust Medical Image Reconstruction

Medical Imaging Techniques Such As Magnetic Resonance Imaging And Computed Tomography Play A Vital Role In Disease Diagnosis And Treatment Planning. However, Medical Images Are Often Degraded By Noise Caused By Low Radiation Exposure, Patient Motion,... Medical Imaging Techniques Such As Magnetic Resonance Imaging And Computed Tomography Play A Vital Role In Disease Diagnosis And Treatment Planning. However, Medical Images Are Often Degraded By Noise Caused By Low Radiation Exposure, Patient Motion, Hardware Limitations, And Quantum Interference. Such Degradation Reduces Image Clarity And May Affect Clinical Decision Making. Conventional Denoising Approaches, Including Wavelet Transforms, Non Local Means Filtering, And Total Variation Minimization, Frequently Struggle To Remove Noise While Preserving Delicate Anatomical Structures. This Paper Presents A Hybrid Framework That Integrates Generative Adversarial Networks, Iterative Reconstruction Algorithms, And Advanced Generative Artificial Intelligence For Effective Medical Image Denoising. The Generative Adversarial Component Models Complex Noise Patterns To Produce Preliminary Clean Images, While The Iterative Reconstruction Process Refines Structural Details Through Optimization. In Addition, Semantic Aware Enhancement Improves Contextual Understanding Of Anatomical Features. Experimental Evaluation On Publicly Available Medical Imaging Datasets Demonstrates Superior Performance In Terms Of Peak Signal To Noise Ratio, Structural Similarity Index, And Mean Squared Error. The Proposed Framework Enhances Diagnostic Reliability And Supports Improved Outcomes In Radiology, Oncology, And Telemedicine Applications. Read more Read less

An Integrated Real-Time Road Surveillance System Using YOLOv8 For Multi-Class Object Detection

The Growing Number Of Road Accidents Worldwide Necessitates The Deployment Of Intelligent Systems Capable Of Continuously Monitoring Roadway Conditions. This Paper Introduces A Comprehensive Computer Vision Framework Designed To Detect Various Roadwa... The Growing Number Of Road Accidents Worldwide Necessitates The Deployment Of Intelligent Systems Capable Of Continuously Monitoring Roadway Conditions. This Paper Introduces A Comprehensive Computer Vision Framework Designed To Detect Various Roadway Features, Including Surface Irregularities, Traffic Control Devices, Elevation Changes, And Biological Entities. The System Utilizes The YOLOv8 Deep Learning Architecture, Which Is Recognized For Its Superior Performance In Balancing Detection Accuracy With Processing Speed. A Specialized Image Dataset Was Curated And Annotated To Train The Model For Recognizing Four Distinct Categories. The Trained System Processes Both Still Images And Continuous Video Feeds, Marking Identified Objects With Bounding Boxes And Confidence Values. Experimental Evaluation Demonstrates That The Proposed Framework Achieves A Mean Average Precision Of 0.93 While Maintaining Real-time Processing Capabilities. This Work Presents A Significant Advancement Over Conventional Approaches That Typically Address Only Single Object Categories, Offering A More Complete Solution For Automated Road Monitoring Applications. Read more Read less

PG-TAF: Policy Governed Tool Access Framework

Modern Enterprises Rely On A Growing Number Of Internal Tools, APIs, And Automated Agents To Perform Critical Operations. However, Direct Tool Access Models Often Result In Fragmented Authorization Logic, Embedded Credentials, Poor Auditability, And... Modern Enterprises Rely On A Growing Number Of Internal Tools, APIs, And Automated Agents To Perform Critical Operations. However, Direct Tool Access Models Often Result In Fragmented Authorization Logic, Embedded Credentials, Poor Auditability, And Limited Governance. These Challenges Increase Security Risks And Operational Complexity, Particularly In Multi-tenant Environments [10]. This Paper Presents The Policy Governed Tool Access Framework (PG-TAF), A Centralized Control Framework Designed To Enforce Secure, Policy-driven Access To Tools And APIs. PG-TAF Introduces A Gateway-based Architecture Where All Tool Requests Are Evaluated Against Configurable Authorization Policies Before Execution [8]. The Framework Integrates Identity Management, Role-based And Attribute-based Access Control (RBAC And ABAC) [1], [3], Centralized Secret Handling With Runtime Injection [9], Lifecycle Enforcement, And Immutable Audit Logging. Unlike Traditional Direct-access Systems, PG-TAF Enables Instant Secret Rotation Without Client-side Updates And Provides Explainable Policy Decisions With Organizational And Workspace-level Precedence Rules. The System Follows A Microservices Architecture To Ensure Separation Of Concerns, Scalability, And Tenant Isolation [7]. A Controlled Evaluation Is Conducted To Analyze Authorization Latency Overhead, Operational Efficiency In Secret Rotation, And Policy Enforcement Correctness Across Multiple Scenarios. Results Demonstrate That PG-TAF Introduces Predictable And Bounded Performance Overhead While Significantly Improving Governance, Auditability, And Operational Security. The Proposed Framework Provides A Practical Foundation For Enterprise-grade Tool Orchestration And Secure API Governance In Modern Distributed Environments. Read more Read less

A State-of-the-Art Review On Seismic Retrofitting And Performance Enhancement Of Reinforced Concrete Structures

Earthquake Around The World Is Single-handedly Responsible For The Destruction To Life And Property In Large Numbers. In Order To Mitigate Such Hazards, It Is Important To Incorporate Norms That Will Enhance The Seismic Performance Of Structures. Thi... Earthquake Around The World Is Single-handedly Responsible For The Destruction To Life And Property In Large Numbers. In Order To Mitigate Such Hazards, It Is Important To Incorporate Norms That Will Enhance The Seismic Performance Of Structures. This Paper Represents The Change Of Reinforced Concrete Structural Components Which Are Found To Exhibit Distress Because Of Earthquake Loading . Such Unserviceable Structures Require Immediate Attention. And It Was Done By Using The Shear Wall Mechanism In The Software .It Can Be Used As A Seismic Retrofitting Technique Because It Can Be Applied Quickly To The Surface Of The Damaged Element Without The Requirement Of Any Special Bonding Material And Also It Requires Less Skilled Labor, As Compared To Other Retrofitting Solutions Presently Existing. It Was Determined That Load Carrying Capacity For Beam-column Joint Retrofitted With Shear Wall Is Increased. In This Paper We Use Analytical Approach. In This We Use Stadd Pro V8i Software. Read more Read less

Design And Development Of Fluid Frictionless Clutch For Vehicles

Friction Clutches Are Vital Mechanical Components That Transmit Torque And Power Between Two Rotating Shafts. They Are Widely Used In Various Applications Including Automobiles, Industrial Machinery, And Power Tools. Traditional Friction Clutches Ope... Friction Clutches Are Vital Mechanical Components That Transmit Torque And Power Between Two Rotating Shafts. They Are Widely Used In Various Applications Including Automobiles, Industrial Machinery, And Power Tools. Traditional Friction Clutches Operate On The Principle Of Frictional Force, Where Two Surfaces Are Pressed Together To Transfer Torque. This Mechanism, However, Leads To Significant Torque Losses, Wear And Tear Of Clutch Plates, Maintenance Issues, And Thermal Heating Of Engine Components. This Paper Presents The Design, Development, And Experimental Evaluation Of A Fluid Frictionless Clutch System That Replaces Mechanical Contact With Fluid-based Torque Transmission Using Electromagnetic And Viscous Fluid Principles. The Proposed System Employs A 2 Mm Oil Gap Between Drive And Driven Plates, Eliminating Direct Surface Contact And Thereby Reducing Friction To Near-zero. The System Significantly Reduces Vibration, Noise, And Maintenance Requirements While Improving Clutch Life And Power Transmission Efficiency. Experimental Results Confirm Smooth Clutch Operation, Negligible Vibration Levels, And Improved Torque Transfer Characteristics. Read more Read less

Intelligent Multi-Sensor Security And Hazard Detection System Using IoT

Modern Residential And Industrial Environments Require Intelligent Monitoring Systems Capable Of Identifying Security Threats And Hazardous Conditions In Real Time. Conventional Safety Systems Often Rely On Isolated Sensors Or Manual Supervision, Whi... Modern Residential And Industrial Environments Require Intelligent Monitoring Systems Capable Of Identifying Security Threats And Hazardous Conditions In Real Time. Conventional Safety Systems Often Rely On Isolated Sensors Or Manual Supervision, Which May Not Provide Comprehensive Protection Against Multiple Environmental Risks. This Work Presents An Intelligent Multi-sensor Security And Hazard Detection System Based On Internet Of Things (IoT) TechnologyThe Proposed System Integrates Temperature Sensors, Gas Sensors, PIR Motion Sensors, And Electrical Fault Detection Modules To Continuously Monitor Environmental Conditions And Security Status.The Collected Sensor Data Are Processed Through A Microcontroller-based Platform And Transmitted Through An IoT Communication Network For Real-time Monitoring. When Abnormal Conditions Such As Gas Leakage, High Temperature, Unauthorized Motion, Or Electrical Faults Are Detected, The System Automatically Triggers Alarms And Sends Notifications To Users. This Enables Rapid Response And Preventive Safety Measures. The Proposed System Offers A Cost-effective, Scalable, And Reliable Monitoring Solution Suitable For Smart Homes, Industrial Environments, And Laboratories. Experimental Implementation Demonstrates That The System Improves Safety Awareness And Enhances Early Hazard Detection Through Integrated Multi-sensor Monitoring.. Read more Read less

Smart Waste Management And Clean Optimization Systems

Rapid Urbanization, Population Growth, And Industrial Expansion Have Led To A Critical Escalation In Municipal Waste Generation, Overwhelming Traditional Management Systems. Conventional Approaches, Which Depend On Static Collection Schedules And Fix... Rapid Urbanization, Population Growth, And Industrial Expansion Have Led To A Critical Escalation In Municipal Waste Generation, Overwhelming Traditional Management Systems. Conventional Approaches, Which Depend On Static Collection Schedules And Fixed Routes, Frequently Result In Operational Inefficiencies, Overflow Events, Elevated Fuel Consumption, And Increased Public Health Risks. This Paper Proposes A Smart Waste Management And Clean Optimization System That Integratesenabled Smart Bins With Real-time Monitoring Capabilities. The System Employs Embedded Sensors To Measure Fill Levels, Detect Hazardous Materials, And Track Environmental Parameters Such As Odor And Temperature. Collected Data Is Transmitted To A Centralized Cloud Platform Where Artificial Intelligence (AI) And Machine Learning (ML) Algorithms Perform Predictive Analytics To Dynamically Optimize Collection Schedules And Route Planning. By Transitioning From Reactive To Predictive Operations, The Proposed Framework Aims To Minimize Logistical Costs, Reduce Carbon Emissions, And Enhance Urban Sanitation Standards. The Efficacy Of The System Is Demonstrated Through [simulation Results/case Study Data], Showing Significant Improvements In Operational Efficiency And Resource Allocation Compared To Conventional Methods. Read more Read less

Real-Time Sign Language Interpretation Using Deep Learning

Sign Language Serves As The Primary Mode Of Communication For Millions Of Hearing-impaired Individuals Worldwide, Yet The Gap Between The Deaf Community And The Hearing World Remains A Significant Barrier. This Paper Presents A Comprehensive, Three-s... Sign Language Serves As The Primary Mode Of Communication For Millions Of Hearing-impaired Individuals Worldwide, Yet The Gap Between The Deaf Community And The Hearing World Remains A Significant Barrier. This Paper Presents A Comprehensive, Three-subsystem Deep Learning Framework For Real-Time Sign Language Interpretation, Encompassing Sign-to-Text Conversion, Text-to-Animation Synthesis, And Text-to-Image Generation. The Sign-to-Text Module Leverages A Convolutional Neural Network (CNN) Trained On MediaPipe-extracted Hand Landmark Vectors, Deployed Via A Flask Backend With OpenCV-based Real-time Video Capture, Achieving A Classification Accuracy Of 97.6% Across 26 American Sign Language (ASL) Gestures. The Text-to-Animation Subsystem Utilizes Angular 21 And Ionic 8 On The Frontend With Node.js/Express And Firebase Cloud Infrastructure, Employing MediaPipe Holistic For Body-pose-driven Skeletal Animation Rendering. The Text-to-Image Subsystem Is A React 19 Single-page Application Powered By A Generative Deep Learning Model Pipeline, Converting Descriptive Text Into Contextual Sign Language Visual Representations. Extensive Experiments Demonstrate Strong Cross-subsystem Performance, With MSE Of 0.043, RMSE Of 0.207, Precision Of 96.8%, And Recall Of 97.1% On The Test Set. The Proposed Integrated Framework Significantly Advances Assistive Communication Technology And Establishes A Replicable Architecture For Real-world Sign Language Translation Deployment. Read more Read less

Mini Scada System

Supervisory Control And Data Acquisition (SCADA) Systems Are Widely Used In Industries For Monitoring And Controlling Various Processes In Real Time. However, Traditional SCADA Systems Are Expensive And Complex, Making Them Unsuitable For Small-scale... Supervisory Control And Data Acquisition (SCADA) Systems Are Widely Used In Industries For Monitoring And Controlling Various Processes In Real Time. However, Traditional SCADA Systems Are Expensive And Complex, Making Them Unsuitable For Small-scale Applications And Educational Purposes. This Project Presents The Design And Development Of A Mini SCADA System Using ESP32 And IoT Technology To Monitor Environmental Parameters, Electrical Parameters, And Control Devices Remotely Through A Mobile Application. The Proposed System Is Built Using An ESP32 Microcontroller, Which Acts As The Central Processing Unit. Various Sensors Are Integrated With The System To Collect Real-time Data. A DHT11 Sensor Is Used For Measuring Temperature And Humidity, An MQ135 Gas Sensor Detects Air Pollution Levels, A Soil Moisture Sensor Monitors Soil Water Content, And An Ultrasonic Sensor Measures Water Level. Additionally, An INA219 Sensor Is Used To Monitor Electrical Parameters Such As Voltage, Current, And Power Consumption. Read more Read less

Bio-Enzyme Stabilization Of Expansive Soils Sustainable And Eco-friendly Alternative To Lime/cement

Expansive Soils, Particularly Black Cotton Soils, Exhibit Significant Swelling And Shrinkage Behavior Due To Moisture Variation, Causing Structural Instability And Damage To Infrastructure. Traditional Stabilization Techniques Using Lime And Cement,... Expansive Soils, Particularly Black Cotton Soils, Exhibit Significant Swelling And Shrinkage Behavior Due To Moisture Variation, Causing Structural Instability And Damage To Infrastructure. Traditional Stabilization Techniques Using Lime And Cement, Although Effective, Are Associated With High Carbon Emissions And Environmental Degradation. This Research Paper Explores Bio-enzyme Stabilization As A Sustainable And Eco-friendly Alternative. Bio-enzymes, Derived From Organic Fermentation, Improve Soil Properties Through Catalytic Reactions That Enhance Particle Bonding And Reduce Water Affinity. This Study Reviews Mechanisms, Experimental Findings, Advantages, Limitations, And Field Applications Of Bio-enzyme Stabilization. Results Indicate Substantial Improvements In Strength, Reduction In Plasticity, And Enhanced Durability, Making Bio-enzymes A Promising Alternative For Sustainable Geotechnical Engineering Read more Read less

AI-based Damage Detection Of Buiding Using Drone Imagery

Unmanned Aerial Vehicles (UAVs), Commonly Known As Drones, Equipped With High Resolution Cameras Generate Vast Amounts Of Structural Imagery Useful For Assessing Building Damage After Earthquakes, Storms, Or Explosions. This Paper Presents A Comprehe... Unmanned Aerial Vehicles (UAVs), Commonly Known As Drones, Equipped With High Resolution Cameras Generate Vast Amounts Of Structural Imagery Useful For Assessing Building Damage After Earthquakes, Storms, Or Explosions. This Paper Presents A Comprehensive Framework For Automated Damage Detection Using Deep Learning Models Trained On Drone Imagery. We Combine Computer Vision Pipelines With Convolutional Neural Networks (CNNs) For Accurate Pixel Level And Object Level Damage Classification. Experimental Results On Benchmark Datasets Show High Accuracy, Demonstrating The Effectiveness Of AI Based Methods For Rapid Post Disaster Building Evaluation. Read more Read less

YOLO-BASED AUTOMATED PCB DEFECT DETECTION AND QUALITY ASSESSMENT SYSTEM FOR SMART MANUFACTURING

Printed Circuit Boards (PCBs) Serve As The Structural And Electrical Foundation Of Virtually Every Modern Electronic Device, Making Their Quality And Reliability Paramount In Both Consumer And Industrial Electronics. Defects Introduced During PCB Fab... Printed Circuit Boards (PCBs) Serve As The Structural And Electrical Foundation Of Virtually Every Modern Electronic Device, Making Their Quality And Reliability Paramount In Both Consumer And Industrial Electronics. Defects Introduced During PCB Fabrication And Assembly — Including Missing Holes, Open Circuits, Solder Bridging, Mouse Bites, Spurious Copper Deposition, And Component Misalignment — Can Precipitate Device Failures, Safety Hazards, And Substantial Economic Losses Across The Manufacturing Supply Chain. Traditional Manual Inspection Approaches Are Inadequate For High-throughput Production Due To Their Low Throughput, Susceptibility To Human Error, And Inability To Detect Subtle Microscopic Anomalies. Classical Automated Optical Inspection (AOI) Systems Based On Template Matching Offer Partial Improvements But Remain Brittle Against Illumination Variations, PCB Surface Reflectivity, And Design Diversity. This Paper Presents A Comprehensive YOLOv11-based Automated PCB Defect Detection And Quality Assessment System Specifically Engineered For Smart Manufacturing Environments. The Proposed Framework Encompasses Five Integrated Modules: Dataset Collection, Data Preprocessing With Augmentation, YOLOv11 Model Fine-tuning With Transfer Learning, Real-time Multi-class Defect Detection, And A Post-detection Quality Assessment Pipeline Comprising Defect Classification, Severity Scoring, And Corrective Recommendation Generation. Experimental Evaluation Demonstrates A Detection Accuracy Of 92%, Precision Of 91%, Recall Of 90%, And F1-score Of 90.5%, Substantially Outperforming Manual Verification (60%), Traditional Image Processing (68%), CNN-based Classification (78%), And Semi-supervised Learning Approaches (83%). Real-time Inference Is Achieved At Approximately 22 Ms Per Image On Standard CPU Hardware, Satisfying Industrial Throughput Requirements. Read more Read less

SMART IoT-BASED RAILWAY CROSSING SYSTEM USING WIRELESS COMMUNICATION

Railway Crossing Accidents Are A Major Concern In Many Countries Due To Human Errors And Improper Gate Management. This Paper Presents An IoT-based Smart Railway Crossing System Designed To Improve Safety At Railway Level Crossings. The Proposed Syst... Railway Crossing Accidents Are A Major Concern In Many Countries Due To Human Errors And Improper Gate Management. This Paper Presents An IoT-based Smart Railway Crossing System Designed To Improve Safety At Railway Level Crossings. The Proposed System Uses Sensors To Detect The Arrival Of A Train And Automatically Close The Railway Gate To Prevent Vehicles And Pedestrians From Crossing The Tracks. Technologies Such As IR Sensors, RFID Modules, And A Microcontroller Are Used To Monitor Train Movement And Control The Gate Mechanism. The System Also Integrates An IoT Platform For Real-time Monitoring And Alert Notifications. This Automated System Helps Reduce Accidents, Minimises Human Intervention, And Improves The Efficiency Of Railway Crossing Operations. Read more Read less

SOLAR AND WIND-BASED SMART STREET LIGHTING WITH WIRELESS CHARGING FOR ELECTRIC VEHICLES AND POWER GENERATION USING SPEED BREAKERS

The Project “Solar And Wind-Based Smart Street Lighting With Wireless Charging For EVs And Power Generation Using Speed Breakers” Presents A Hybrid Renewable Energy System That Combines Solar, Wind, And Mechanical Energy Sources. A 12V Solar Panel Wi... The Project “Solar And Wind-Based Smart Street Lighting With Wireless Charging For EVs And Power Generation Using Speed Breakers” Presents A Hybrid Renewable Energy System That Combines Solar, Wind, And Mechanical Energy Sources. A 12V Solar Panel With An MPPT Controller Generates Power During The Day, While Wind Turbines Provide Energy At Night Or During Low Sunlight Conditions. Additional Power Is Produced Using Piezoelectric Sensors Installed Under Speed Breakers, Converting Vehicle Pressure Into Electricity. All Generated Energy Is Stored In A Rechargeable Battery And Used For Smart Street Lighting And Wireless EV Charging Through Dynamic Energy Transfer (DET). The System Also Includes An Arduino-based Monitoring Setup With An LCD Display. This Integrated Approach Improves Energy Efficiency, Ensures Continuous Power Supply, And Supports Sustainable Smart City Infrastructure. Read more Read less

IOT BASED FIRE PREVENTION SYSTEM FOR ELECTRIC VEHICLE BATTERIES

Electric Vehicles Require Intelligent Monitoring To Ensure Safety, Efficiency, And Longevity. This Study Proposes An IoT-enabled Battery Monitoring And Fault Detection System Targeting Risks Like Overheating, Voltage Imbalance, And Fire Hazards. The... Electric Vehicles Require Intelligent Monitoring To Ensure Safety, Efficiency, And Longevity. This Study Proposes An IoT-enabled Battery Monitoring And Fault Detection System Targeting Risks Like Overheating, Voltage Imbalance, And Fire Hazards. The Framework Integrates Voltage, Temperature, And Flame Sensors To Capture Real-time Battery Data. An ATmega328P Microcontroller Processes Inputs Against Predefined Thresholds To Detect Anomalies. NodeMCU Enables Wireless Transmission To An IoT Cloud For Remote Visualization And Diagnostics. Upon Fault Detection, The System Triggers LCD Alerts And Relay Isolation To Safeguard The Battery Circuit. This Proactive Design Facilitates Early Intervention, Minimizes Downtime, And Boosts Reliability. Remote Access Empowers Users And Technicians For Informed Maintenance Decisions. Cost-effective And Scalable, The Solution Adapts To Diverse EV Applications. By Merging Embedded Systems With IoT, It Advances Battery Management, Extends Lifespan, And Reduces Fire Risks. Future Enhancements Include Predictive Analytics, Mobile Integration, And Data Logging For Sustainable Energy Utilization. Read more Read less

Landslide Alert System Using Machine Learning And IoT Sensors For Real Time Prediction And Notification

Natural Disasters, Such As Heavy Rainfall And Landslides, Pose Significant Threats To Life And Infrastructure, Particularly In Vulnerable Regions. This Work Proposes A Hybrid Disaster Alert System That Utilizes An IoT-enabled Embedded System, Integra... Natural Disasters, Such As Heavy Rainfall And Landslides, Pose Significant Threats To Life And Infrastructure, Particularly In Vulnerable Regions. This Work Proposes A Hybrid Disaster Alert System That Utilizes An IoT-enabled Embedded System, Integrating Various Environmental Sensors To Predict And Alert Users About Potential Hazards. Landslides Are A Significant Natural Hazard That Can Cause Extensive Damage To Infrastructure And Loss Of Life. Timely And Accurate Prediction Of Landslides Is Crucial For Mitigating These Risks. This Work Presents An Advanced Landslide Alert System Leveraging Deep Learning Techniques And Internet Of Things (IoT) Sensors For Real-time Prediction And Notification. The System Utilizes A Recurrent Neural Network (RNN) Model Trained On Historical Landslide Data And Real-time Sensor Inputs, Including Soil Moisture, Temperature, Humidity, And Ground Vibration. These Sensors, Interfaced With An Arduino Microcontroller, Continuously Monitor The Environmental Conditions And Transmit Data To The RNN Model. The Model Processes This Data To Predict The Likelihood Of A Landslide And Triggers An Alert If The Risk Exceeds A Predefined Threshold. The System's Architecture Ensures Low Latency And High Accuracy In Predictions, Enabling Timely Evacuation And Preventive Measures. This Integrated Approach Combines The Power Of Deep Learning With IoT Technology To Provide A Robust And Reliable Landslide Early Warning System, Significantly Enhancing Disaster Preparedness And Response Capabilities. Read more Read less

DEEP VISION-BASED SMART WASTE SORTING USING VGG16 AND YOLO FOR REAL-TIME APPLICATIONS

Waste Classification Is A Critical Component Of Effective Waste Management, As It Enables Materials To Be Properly Segregated For Disposal, Recycling, And Environmental Protection. Traditional Waste Sorting Methods Largely Depend On Manual Labor, Whi... Waste Classification Is A Critical Component Of Effective Waste Management, As It Enables Materials To Be Properly Segregated For Disposal, Recycling, And Environmental Protection. Traditional Waste Sorting Methods Largely Depend On Manual Labor, Which Is Not Only Time-consuming But Also Prone To Errors, Leading To Contamination Of Recyclable Materials And Inefficiencies In Waste Processing. The Rapid Increase In Global Waste Generation Has Highlighted The Urgent Need For Automated, Accurate, And Scalable Solutions. This Project Proposes A Smart Waste Classification System That Leverages Deep Learning Techniques, Specifically The VGG16 Convolutional Neural Network (CNN) Architecture, To Automatically Categorize Waste Materials Into Classes Such As Paper, Plastic, Glass, Metal, And Organic Waste. The System Uses Extensive Image Preprocessing And Augmentation Techniques To Enhance Model Performance And Robustness. In Addition, The Project Integrates The YOLO (You Only Look Once) Framework For Real-time Object Detection, Allowing The System To Efficiently Recognize And Classify Waste Items In Dynamic Environments. By Combining VGG16 For Feature Extraction And YOLO For Detection, The System Significantly Reduces Human Intervention, Improves Sorting Accuracy, And Accelerates The Recycling Process Read more Read less

DEEPTRACENET: AN AI DRIVEN FRAMEWORK FOR MATHEMATICAL FEATURE-BASED FORGERY DETECTION

The Rapid Growth Of Deepfake Technologies Has Raised Serious Concerns Regarding The Authenticity Of Digital Images And Videos, Impacting Information Security, Media Credibility, And Forensic Analysis. This Research Presents A Deep Learning-based Fram... The Rapid Growth Of Deepfake Technologies Has Raised Serious Concerns Regarding The Authenticity Of Digital Images And Videos, Impacting Information Security, Media Credibility, And Forensic Analysis. This Research Presents A Deep Learning-based Framework For Identifying AI-generated Manipulated Content With Improved Accuracy And Efficiency. A Hybrid Convolutional Neural Network (CNN) Architecture Is Employed, Where ResNet Serves As The Backbone For Extracting High-level Spatial Features From Input Media. The Model Is Enhanced With Inverted Residual Blocks And Linear Bottlenecks To Preserve Essential Feature Information While Reducing Computational Complexity And Memory Usage. The Proposed Approach Enables Automatic Learning Of Discriminative Features, Eliminating The Dependence On Manual Feature Engineering. Input Images And Video Frames Undergo Preprocessing Followed By Feature Extraction And Classification Stages To Determine Authenticity. The Integration Of Advanced Architectural Components Contributes To Faster Inference And Better Generalization Across Diverse Deepfake Manipulation Techniques. Experimental Considerations Indicate That The Framework Is Capable Of Effectively Distinguishing Between Real And Forged Content, Even When Visual Differences Are Minimal. Overall, This Research Provides A Robust And Scalable Solution For Deepfake Detection, Ensuring Reliable Identification Of Manipulated Media And Supporting The Preservation Of Digital Content Integrity. Read more Read less

A Personal And Family Financial Decision Support System

Managing Personal And Family Finances Has Become More Challenging In Recent Years Due To Increasing Living Expenses And The Presence Of Multiple Sources Of Income. The Personal And Family Financial Decision Support System Is Developed To Simplify Thi... Managing Personal And Family Finances Has Become More Challenging In Recent Years Due To Increasing Living Expenses And The Presence Of Multiple Sources Of Income. The Personal And Family Financial Decision Support System Is Developed To Simplify This Process By Offering A Single, Convenient Platform Where Users Can Track Their Expenses, Plan Their Savings, And Make More Informed Financial Decisions.By Combining Budgeting Tools, Financial Tracking, And Data Visualization, It Turns Raw Financial Data Into Useful Information.These Information Helps Users In Managing Their Income, Expenses, And Investments Better. With Careful Analysis And Organized Data, It Helps Families Improve Their Financial Planning, Reduce Unnecessary Spending, And Promote Long-term Financial Security. Read more Read less

A Study On The Perception Of Plant-Based Protein Products Among Consumers : Survey-based Analysis

Nowadays, Plant-based Protein Products Are In Market Demand And Has Gained Attention Due To Their Nutritional Content, Health Improvements And Sustainability. With The Growing Demand For Plant-based Protein, Analyzing And Understanding Consumer Aware... Nowadays, Plant-based Protein Products Are In Market Demand And Has Gained Attention Due To Their Nutritional Content, Health Improvements And Sustainability. With The Growing Demand For Plant-based Protein, Analyzing And Understanding Consumer Awareness And Preference Is Essential For Improving Food Production Techniques And Support Systems. The Objective Of This Study Was To Investigate Consumer Awareness, Perception And Consumption Of Plant-based Protein. An Online Survey Questionnaire Was Prepared And Shared To Collect Data On Demographic Information, Type Of Diet, Knowledge Of Plant-based Protein, Experience, Willingness To Buy, Taste Preference And Price Ranges.The Collected Data Were Collected From 161 Participants And The Findings Shows That Majority Of Participants Were Aware Of Plant-based Protein And Has Positive Perception Due To Its Health Benefits And Environmental Concerns. However, Purchasing Behaviour Was Affected By Product Variety, Availability, Taste Preference And Higher Price Ranges. Around 80.7% Of Participants Showed Willingness To Increase Their Consumption Of Plant-based Protein In The Future. The Study Reveals That, Providing Awareness, Ensuring Availability And Affordability Will Increase Market Growth For Plant-based Protein Read more Read less

Smart Review Analysis System Using Machine Learning

The Rapid Proliferation Of Online Movie Reviews Has Created A Pressing Need For Automated Systems Capable Of Distinguishing Spoiler Content From Non-spoiler Opinions. This Paper Presents The Smart Review Analysis System (SRAS), An Intelligent Spoiler... The Rapid Proliferation Of Online Movie Reviews Has Created A Pressing Need For Automated Systems Capable Of Distinguishing Spoiler Content From Non-spoiler Opinions. This Paper Presents The Smart Review Analysis System (SRAS), An Intelligent Spoiler Detection Framework Built Upon A Bidirectional Long Short-Term Memory (Bi-LSTM) Neural Network Augmented With Pre-trained GloVe Word Embeddings. The Proposed System Processes IMDB User Reviews And Classifies Them As Spoiler Or Non-spoiler With High Accuracy. Extensive Preprocessing, Tokenization, And Sequence Padding Are Applied To The Textual Data Prior To Model Training. The Architecture Employs Stacked Bi-LSTM Layers, Spatial Dropout For Regularization, And A Sigmoid Output Layer For Binary Classification. Experimental Results On The IMDB Spoiler Dataset Demonstrate That SRAS Achieves Competitive Classification Performance, Validated Through Accuracy, Precision, Recall, And F1-score Metrics. The System Provides A Practical And Scalable Solution For Real-time Spoiler Filtering In Movie Review Platforms, Enhancing User Experience And Content Discovery. Read more Read less

Impact Of Quick-Commerce (Blinkit, Zepto) On Urban Consumer Behaviour

This Study Investigates The Impact Of Quick-commerce Platforms — Specifically Blinkit And Zepto — On Urban Consumer Behavior In India. With The Rapid Proliferation Of Ultra-fast Delivery Services Promising Fulfilment Within 10 To 20 Minutes, Underst... This Study Investigates The Impact Of Quick-commerce Platforms — Specifically Blinkit And Zepto — On Urban Consumer Behavior In India. With The Rapid Proliferation Of Ultra-fast Delivery Services Promising Fulfilment Within 10 To 20 Minutes, Understanding How These Platforms Reshape Purchase Patterns, Brand Loyalty, Impulse Buying, And Consumer Satisfaction Has Become Increasingly Critical For Marketers And Retailers. A Structured Questionnaire-based Survey Was Administered To 188 Urban Respondents, And The Data Were Analysed Using Pearson Correlation Analysis And Multiple Linear Regression. The Findings Reveal That Convenience And Speed Perception Significantly And Positively Influences Both Impulse Buying Behavior And Brand Switching Tendency. Consumer Satisfaction Is Most Strongly Driven By Convenience And Speed Perception, While Impulse Buying Behavior Also Demonstrates A Significant Direct Effect. Brand Switching Tendency Does Not Independently Predict Satisfaction When Other Variables Are Controlled. These Results Offer Actionable Insights For Q-commerce Platforms And Traditional Retailers Navigating The Competitive Urban Essentials Market. Read more Read less

Traffic Signal Violation Detection System

Managing Traffic At Busy Intersections Has Become Increasingly Challenging Due To Frequent Signal Violations By Drivers. Ignoring Red Signals Often Leads To Unsafe Road Conditions And Traffic Disturbances. To Address This Issue, This Project Proposes... Managing Traffic At Busy Intersections Has Become Increasingly Challenging Due To Frequent Signal Violations By Drivers. Ignoring Red Signals Often Leads To Unsafe Road Conditions And Traffic Disturbances. To Address This Issue, This Project Proposes An Automated Monitoring System That Observes Vehicle Movement Near Traffic Signals Without Continuous Human Supervision. A Key Feature Of This System Is Its Ability To Support Emergency Ambulances. When An Ambulance Approaches The Signal, The System Detects It And Gives Immediate Priority By Allowing It To Pass Without Delay. This Reduces Waiting Time In Critical Situations And Improves Emergency Response Efficiency. Read more Read less

LOAN APPROVAL SYSTEM USING MACHINE LEARNING SCORING

This Paper Presents LoanAI, A Machine Learning-based Loan Approval System Designed To Address Financial Exclusion Among Young Borrowers Lacking Traditional Credit History. The System Evaluates Creditworthiness Using Bank Transaction Data Such As Sala... This Paper Presents LoanAI, A Machine Learning-based Loan Approval System Designed To Address Financial Exclusion Among Young Borrowers Lacking Traditional Credit History. The System Evaluates Creditworthiness Using Bank Transaction Data Such As Salary Consistency, Savings Patterns, And Cash Dependency. An XGBoost Classifier Is Used To Generate A Simulated CIBIL Score Ranging From 300 To 850. A Fraud Detection Mechanism Identifies Anomalies Including High FOIR And Irregular Transaction Patterns. The System Processes Loan Applications In Under 3 Seconds And Improves Approval Rates From 40% To 65% While Maintaining A Low Default Rate Of 4.2%. The Proposed Solution Demonstrates 82% Prediction Accuracy And Enables Scalable, Real-time Loan Decision-making. Read more Read less

COMPARATIVE AND STAGE-WISE ALZHEIMER’S DISEASE PREDICTION USING HYBRID INTELLIGENCE

In This Paper, Alzheimer’s Disease (AD) Is A Progressive Neurodegenerative Disorder Characterized By Memory Loss And Cognitive Decline. Accurate And Early Diagnosis Is Critical To Intervene Effectively, Yet Reliable And Accurate Diagnosis Is Difficul... In This Paper, Alzheimer’s Disease (AD) Is A Progressive Neurodegenerative Disorder Characterized By Memory Loss And Cognitive Decline. Accurate And Early Diagnosis Is Critical To Intervene Effectively, Yet Reliable And Accurate Diagnosis Is Difficult Because Of The Intricate Relationship Between The Cognitive Symptoms And The Neurological Changes. The Paper Suggests A Hybrid Intelligence Model For Stage-wise Alzheimer’s Disease Prediction Using Cognitive Assessment, Speech Analysis, And Machine Learning Techniques. The Proposed Model Is Composed Of Three Analysis Modules. The First Module Is The Cognitive Assessment Analysis, Where The Data Collected Through Memory-based Tasks, Number Sequence, Visual Recognition, And Questionnaires Are Analyzed Using Machine Learning Techniques To Estimate The Preliminary Cognitive Risk Levels. The Second Module Is The Speech Analysis, Where The Data Collected Through Speech And Language Analysis Are Processed Using Natural Language Processing Techniques To Evaluate The Verbal Fluency, Recall, And Speech Patterns Related To Cognitive Deterioration. The Third Module Is The Multimodal Fusion Analysis, Where The Predictions From The Cognitive And Speech Analysis Are Fused To Generate The Stage-wise Classification Of Alzheimer’s Disease, Ranging From Normal, Mild Cognitive Impairment, And Advanced Stages. Proposed Model Also Considers The Application Of Explainable Artificial Intelligence Techniques To Generate Interpretable Outputs, Facilitating Better Understanding And Decision-making Among The Medical And Caregiving Communities. The Experimental Results Show That The Proposed Model Provides Improved Accuracy And Consistency Compared To The Single-modality Assessment Techniques. Read more Read less

Smart Industrial Product Counting And Sorting Machine

In Modern Industrial Automation, Accuracy, Speed, And Reliability In Production Line Monitoring Are Critical Factors For Improving Productivity And Minimizing Human Intervention. The Smart Industrial Product Counting And Sorting System Is Designed To... In Modern Industrial Automation, Accuracy, Speed, And Reliability In Production Line Monitoring Are Critical Factors For Improving Productivity And Minimizing Human Intervention. The Smart Industrial Product Counting And Sorting System Is Designed To Automate The Process Of Counting Products And Separating Defective Items Using Embedded Systems And Sensor-based Technology. This Project Integrates Mechanical And Electronic Subsystems To Create An Efficient, Low-cost, And User-friendly Industrial Solution. In Conclusion, The Smart Industrial Product Counting And Sorting System Provides A Compact, Cost-effective, And Efficient Solution For Modern Production Lines, Aligning With The Growing Demand For Automation In Industries. Read more Read less

Traditional Grain-Based Foods In Contemporary Diets: A Study On Consumption Frequency

Traditional Grain-based Foods, Such As Rice, Wheat, Millets, Barley And Sorghum, Play A Crucial Role In Nutritional Security And Cultural Dietary Practices. Assessing The Frequency And Quantity Of Traditional Grain Consumption Across Demographics Wil... Traditional Grain-based Foods, Such As Rice, Wheat, Millets, Barley And Sorghum, Play A Crucial Role In Nutritional Security And Cultural Dietary Practices. Assessing The Frequency And Quantity Of Traditional Grain Consumption Across Demographics Will Be Of Prime Significance, Which Would Help In Developing New Food Products. The Present Study Evaluated Consumer Awareness, Consumption Patterns, Accessibility And Perceptions Regarding Traditional Grain-based Foods. A Cross-sectional Survey Was Conducted Among 216 Respondents From Diverse Demographic Backgrounds. The Results Indicated That There Is A Regular Consumption Of Traditional Grains, Like Rice (90.6%), Being The Most Commonly Consumed, Followed By Wheat (6.6%)and Millets(2.8%). Despite Strong Awareness Of Its Nutritional And Cultural Value, Actual Consumption Is Influenced By Taste Preferences, Cost, Availability And Convenience. Addressing These Through Improved Access, Affordable Pricing, Product Innovation (e.g., Ready-to-eat Millet Products)and Targeted Nutrition Education—especially For Older And Lower-income Groups Can Promote Sustained Consumption Of Traditional Grains In Modern Diets. Read more Read less

OWNER PERCEPTION TOWARDS CLIENT-RELATED CHALLENGES IN FITNESS CENTRES IN COIMBATORE CITY

The Fitness Industry Has Witnessed Rapid Growth In Recent Years, Particularly In Urban Regions, Driven By Increasing Awareness Of Health And Lifestyle Management. Despite This Expansion, Fitness Centre Owners Continue To Encounter Significant Challen... The Fitness Industry Has Witnessed Rapid Growth In Recent Years, Particularly In Urban Regions, Driven By Increasing Awareness Of Health And Lifestyle Management. Despite This Expansion, Fitness Centre Owners Continue To Encounter Significant Challenges Related To Client Behaviour. This Study Aims To Analyse The Perception Of Fitness Centre Owners Regarding Client-related Issues Such As Irregular Attendance, Lack Of Commitment, Unrealistic Expectations, And Early Dropout. The Research Is Based On Primary Data Collected From 50 Fitness Centre Owners In Coimbatore City Using A Structured Questionnaire. Statistical Tools Such As Percentage Analysis, ANOVA, Ranking, And Chi-square Tests Were Applied To Interpret The Data. The Findings Indicate That Client Behaviour Remains A Critical Concern Affecting Operational Efficiency And Long-term Sustainability. The Study Concludes That Improving Client Awareness, Motivation, And Engagement Strategies Can Significantly Enhance Retention And Performance Of Fitness Centres. Read more Read less

BLOCKCHAIN-ENABLED HUMAN ORGAN DONATION AND TRANSPLANTATION PORTAL FOR SECURE AND TRANSPARENT MANAGEMENT

The Efficiency Of Organ Transplantation Is Frequently Hindered By Fragmented Data Management And The Lack Of A Centralized Infrastructure For Real-time Donor-recipient Matching. This Paper Proposes A Web-based Organ Donation Management System Designe... The Efficiency Of Organ Transplantation Is Frequently Hindered By Fragmented Data Management And The Lack Of A Centralized Infrastructure For Real-time Donor-recipient Matching. This Paper Proposes A Web-based Organ Donation Management System Designed To Bridge This Gap By Digitizing The End-to-end Donation Workflow. The System Provides A Unified Platform For Donor Registration And Clinical Verification, Significantly Reducing The Latency Involved In Identifying Compatible Donors During Medical Emergencies. By Replacing Traditional, Paper-based Record-keeping With A Centralized Digital Repository, The Architecture Eliminates Data Redundancy And Minimizes Human Administrative Errors. To Address The Critical Requirement For Patient Privacy And Data Security, The Platform Integrates Blockchain Technology. This Ensures That All Donor And Organ-related Records Are Stored In A Decentralized, Tamper-proof Ledger, Fostering Transparency And Preventing Unauthorized Data Manipulation. The Result Is A Secure, Scalable Framework That Enhances Coordination Among Healthcare Stakeholders And Optimizes The Decision-making Process In Organ Transplantation. Read more Read less

From Centralization To Decentralization: Blockchain's Role In Transforming Social Media Platform Page

The Growing Dependence On Social Media Platforms For Day-to-day Communication Has Made Digital Privacy A Serious Concern. Platforms Such As Instagram, Twitter, And Facebook Host Vast Amounts Of User-generated Content, Including Images, Text, And Meta... The Growing Dependence On Social Media Platforms For Day-to-day Communication Has Made Digital Privacy A Serious Concern. Platforms Such As Instagram, Twitter, And Facebook Host Vast Amounts Of User-generated Content, Including Images, Text, And Metadata, Much Of Which Can Unintentionally Expose Personal Details. Despite The Available Privacy Controls, These Tools Often Fall Short Of Preventing The Misuse Or Unauthorized Distribution Of Sensitive Content. This Study Presents A Secure Image-sharing Framework That Combines Wavelet-SVD-based Watermarking, Steganography, And Blockchain Technology To Address These Vulnerabilities. The System Supports Image Categorization, Invisible Watermark Embedding, And Screen-shot Prevention. Blockchain Ledgers Ensure Tamper-proof Storage Of Image Ownership Records, Whereas The Dual-layered Security Approach Makes The Unauthorized Extraction Of Hidden Content Extremely Difficult. Read more Read less

A Study On Consumer Preference For Brick -and -mortar Fashion Retail Brands In Coimbatore City

This Study Analyzes Consumer Preferences Towards Brick-and-mortar Fashion Retail Brands In Coimbatore City. It Focuses On Factors Influencing Buying Behavior Such As Quality, Price, Store Ambience, And Product Variety. A Descriptive Research Design W... This Study Analyzes Consumer Preferences Towards Brick-and-mortar Fashion Retail Brands In Coimbatore City. It Focuses On Factors Influencing Buying Behavior Such As Quality, Price, Store Ambience, And Product Variety. A Descriptive Research Design Was Adopted, With Data Collected From 97 Respondents Using Structured Questionnaires. Statistical Tools Like Percentage Analysis, Chi-square, And ANOVA Were Used For Analysis. The Findings Indicate That Quality And In-store Experience Are The Key Drivers Of Consumer Choice. Demographic Factors Such As Age, Income, And Occupation Also Influence Purchasing Decisions. The Study Concludes That Physical Retail Stores Remain Relevant, And Improving Customer Experience And Promotional Strategies Can Enhance Satisfaction And Loyalty. Read more Read less

Plant Disease Detection Using Deep Learning

Agriculture Plays A Crucial Role In Global Food Security, And Plant Diseases Represent One Of The Most Significant Threats To Crop Productivity And Quality Worldwide. Traditional Methods Of Disease Identification Rely Heavily On Manual Inspection By... Agriculture Plays A Crucial Role In Global Food Security, And Plant Diseases Represent One Of The Most Significant Threats To Crop Productivity And Quality Worldwide. Traditional Methods Of Disease Identification Rely Heavily On Manual Inspection By Agricultural Experts, Which Is Time-consuming, Costly, And Prone To Human Error. To Overcome These Limitations, This Work Proposes A Deep Learning-based Plant Disease Detection System Capable Of Automatically Identifying And Classifying Diseases From Leaf Images. The Proposed Framework Employs Convolutional Neural Networks (CNN) To Extract Discriminative Features From Plant Leaf Images And Performs Multi-class Disease Classification With High Accuracy. Transfer Learning Techniques Using Pre-trained Models Such As ResNet And VGG Are Incorporated To Improve Generalization And Reduce Training Time. The System Enables Early And Accurate Disease Diagnosis, Allowing Farmers To Take Timely Corrective Action And Minimize Crop Losses. Experimental Results Demonstrate That The Proposed Model Achieves Competitive Accuracy On Benchmark Plant Disease Datasets, Making It A Reliable Tool For Precision Agriculture Applications. Read more Read less

Alumni Connect-A Platform For Alumni Engagement And Student Development

The Alumni Connect Platform Is A Modern, Web-enabled System Built To Strengthen Bonds Between Former And Current Students, Raising The Likelihood Of Meaningful Employment Connections While Encouraging Alumni Participation Within Their Academic Commun... The Alumni Connect Platform Is A Modern, Web-enabled System Built To Strengthen Bonds Between Former And Current Students, Raising The Likelihood Of Meaningful Employment Connections While Encouraging Alumni Participation Within Their Academic Communities. Conventional Alumni Systems Rely Heavily On Manual Processes, Lack Robust Security, And Offer No Intelligent Talent-matching Capabilities. The Alumni Connect Application Addresses These Gaps Through A User-friendly Interface Featuring Biometric Login Using The Grassmann-based Facial Recognition Algorithm, A Centralized Admin Module For Managing Records And Events, And An NLP-driven Job Recommendation Engine That Aligns Student Skills With Available Opportunities. Alumni Can Maintain Professional Profiles, Post Job Vacancies, And Participate In Mentorship Programs, While Students Can Explore Alumni Networks, Upload Their Resumes, And Receive Personalized Career Guidance. The Platform Builds A Structured, Secure, And Intelligent Bridge Between Alumni And Students, Significantly Improving Networking, Career Placement, And Institutional Engagement. Read more Read less

A Study On The Increasing Preference For Online Shopping Over In-Store Shopping

The Rapid Advancement Of Digital Technology And The Widespread Use Of The Internet Have Significantly Transformed Consumer Shopping Behavior. This Study Examines The Increasing Preference For Online Shopping Over Traditional In-store Shopping. It Foc... The Rapid Advancement Of Digital Technology And The Widespread Use Of The Internet Have Significantly Transformed Consumer Shopping Behavior. This Study Examines The Increasing Preference For Online Shopping Over Traditional In-store Shopping. It Focuses On Key Factors Such As Convenience, Trust, Perceived Risk, And Lifestyle Changes Influencing Consumer Decisions. The Research Also Compares Customer Experiences Across Both Shopping Modes And Identifies Reasons For The Decline In In-store Shopping. Using Statistical Tools Such As Percentage Analysis, Correlation, Regression, ANOVA, And Chi-square Tests, The Study Provides Insights Into Evolving Consumer Preferences In The Digital Retail Environment. Read more Read less

Secure Trade Mediation System

In This Modern Era, We Have Faced A Lot Of E-commerce Scams, So We Have Introduced A System To Reduce The This Scam. Wehave Built A Dual Escrow Mediation Where Both The Buyer And Seller Have To Initiate 10% Of The Deposit Inside The Escrow Wallet For... In This Modern Era, We Have Faced A Lot Of E-commerce Scams, So We Have Introduced A System To Reduce The This Scam. Wehave Built A Dual Escrow Mediation Where Both The Buyer And Seller Have To Initiate 10% Of The Deposit Inside The Escrow Wallet For Cash On Delivery For The Prepaid, The Buyer Has To Pay The Whole Amount Price, And That Price Will Be Locked In The Escrow. The 10% Of The Deposit, Which Is Called The Security Money, Will Be Deposited By The Seller Into The Escrow Wallet. In This System, The Agent Delivery Acts As A Judge, A Genuine Return Request And The Unreasonable Request Will Be Handled By The Agent Delivery. If The Return Request Was Unreasonable By The Buyer, Then A Penalty Will Be Given To The Buyer.If The Agent Judges A Return Request As Unreasonable, Then 10% Of Productprice Is Redistributed To The Three Parties The 3% Of The Amount Goes To The Agent For Delivery As A Service Fees. 7% Of The Product Price Is Transferred To The Seller As A Compensation And Then The Remaining 90% Of The Product Price Is Returned To The Buyer. Because Of The Unreasonable Request, The Buyer's Reputation Score Will Be Reduced To 0.1 Leading To A 9.9 Reputation Score. Whenever The Buyer Or Seller Tries To Scam, Their Reputation Score Will Be Reduced To 0.1 . The Reputation Score Will Help Both The Buyer And Seller To Identify That The Particular Party Have Scammed. The System Was Built Using Python3.11+, Streamlit, FastAPI, Uvicorn, And The Requests Library. The System Results Confirm The Escrow Lock, Dual Escrow Mediation, Penalty Redistribution And Repetition Scoring. Read more Read less

A STUDY ON THE EMERGING PREFERENCE TOWARDS DINKY LIFESTYLE AMONG MILLENNIALS IN COIMBATORE CITY

This Study Examines The Growing Trend Of Dual Income, No Kids Yet (DINKY) Households Among Millennials In Coimbatore City, Focusing On The Factors Influencing Their Decision To Delay Parenthood And Its Impact On Financial And Lifestyle Choices. Drive... This Study Examines The Growing Trend Of Dual Income, No Kids Yet (DINKY) Households Among Millennials In Coimbatore City, Focusing On The Factors Influencing Their Decision To Delay Parenthood And Its Impact On Financial And Lifestyle Choices. Driven By Urbanisation, Career Aspirations, Financial Stability, And Changing Social Values, Many Millennial Couples Are Increasingly Adopting The DINKY Lifestyle. The Study Explores How Higher Disposable Income, Evolving Attitudes Towards Marriage And Family, And The Desire For Personal And Professional Growth Shape Their Consumption Patterns, Savings, Investments, And Overall Decision-making Behaviour. By Analysing These Emerging Trends Within The Context Of A Rapidly Developing Tier-two City, The Research Aims To Provide Insights Into Shifting Household Structures And Their Broader Implications For Economic Behaviour And Future Societal Patterns Read more Read less

AI-Powered Hospital Operations & Intelligence Management System

Efficient Hospital Operations Require Structured Data Management And Real-time Monitoring To Support Effective Decision-making. Traditional Hospital Systems Often Suffer From Manual Processes, Fragmented Records, And Inefficient Resource Allocation.... Efficient Hospital Operations Require Structured Data Management And Real-time Monitoring To Support Effective Decision-making. Traditional Hospital Systems Often Suffer From Manual Processes, Fragmented Records, And Inefficient Resource Allocation. This Paper Presents The AI-Powered Hospital Operations & Intelligence Management System (HOIMS), A Centralized Hospital Management Platform Built Using SQL Server, HTML, JavaScript, And Power BI. The System Integrates Patient Admissions, Bed Management, Appointment Scheduling, And Administrative Control Within A Unified Relational Database Framework. Real-time Data Is Visualized Through Interactive Power BI Dashboards To Analyze Bed Occupancy, Patient Trends, And Operational Performance. Data Integrity Is Maintained Using Normalization, Validation Rules, And Primary–foreign Key Constraints. The Proposed System Improves Operational Efficiency, Reduces Manual Errors, And Enhances Transparency. Future Work Includes Integrating Machine Learning Models For Predictive Bed Occupancy And Patient Inflow Forecasting. Read more Read less

Live Cryptocurrency Market Dashboard With Machine Learning-Based Price Prediction And RSI Signal Generator

Cryptocurrency Markets Are Highly Volatile And Change Rapidly, Which Makes Price Prediction And Decision- Making Difficult For Investors. This Project Presents A Live Cryptocurrency Market Dashboard That Provides Real-time Market Visualization Along... Cryptocurrency Markets Are Highly Volatile And Change Rapidly, Which Makes Price Prediction And Decision- Making Difficult For Investors. This Project Presents A Live Cryptocurrency Market Dashboard That Provides Real-time Market Visualization Along With Machine Learning-based Price Prediction And An RSI-based Buy/Sell/Hold Signal System. The System Collects Live Cryptocurrency Data Using The CoinGecko API And Processes It For Analysis. Three Models— LSTM, GRU, And XGBoost—are Used To Predict Future Prices, And Their Performance Is Evaluated Using Metrics Such As MAE, MSE, RMSE, And MAPE. In Addition To Prediction, The Dashboard Calculates The Relative Strength Index (RSI) To Generate Trading Signals That Help Users Understand Whether To Buy, Sell, Or Hold A Cryptocurrency. The Dashboard Is Developed Using Streamlit And Plotly To Provide An Interactive And User-friendly Interface. By Combining Live Data, Machine Learning Models, And Technical Indicators, The Proposed System Acts As A Practical Tool For Cryptocurrency Market Analysis And Basic Investment Decision Support. Read more Read less

An AI-Based Medical ChatBot For Infectious Disease Prediction Using Multi-Layer Perceptron

The Rapid Growth Of Digital Health Technologies Has Created New Opportunities For Accessible And Early Disease Detection. This Project Proposes An Intelligent Chatbot-based System Capable Of Providing Preliminary Disease Diagnosis Using Machine Learn... The Rapid Growth Of Digital Health Technologies Has Created New Opportunities For Accessible And Early Disease Detection. This Project Proposes An Intelligent Chatbot-based System Capable Of Providing Preliminary Disease Diagnosis Using Machine Learning. The Chatbot Interacts With Users Through A Conversational Interface, Collects Symptom Inputs, And Analyzes Them Using Trained ML Models Such As Naïve Bayes, Decision Trees, Or Support Vector Machines. The System Is Designed To Classify Possible Diseases Based On Symptom Patterns And Return Probable Diagnoses Along With Recommended Next Steps, Such As Consulting A Specialist Or Seeking Emergency Care. The Chatbot Also Offers Continuous Guidance, Clarifying Symptoms And Providing Health-awareness Information In Real Time. The Integration Of NLP Enables The System To Understand Natural User Queries, Making It User-friendly And Accessible Even To Individuals With Limited Medical Knowledge. Experimental Results Demonstrate That ML-based Prediction Significantly Improves Diagnostic Accuracy Compared To Rule-based Systems. This Solution Can Support Rural Healthcare, Reduce Clinical Workload, And Provide Immediate Preliminary Medical Assistance, Serving As A Low-cost, Scalable Tool For Early Disease Detection And Decision Support. Read more Read less

A Study On The Effectiveness Of Four- Day Work Week And Employee Awareness Of The Legal Framework Governing It In The IT- Sector

The Four-day Work Week Is A Growing Trend In The Global Workspace And Many Companies Are Trying With This New Model Of Work. India Is Exploring To Implement Four-day Work Week. This Research Paper Examines Of Four-day Work Week In The IT Sector And E... The Four-day Work Week Is A Growing Trend In The Global Workspace And Many Companies Are Trying With This New Model Of Work. India Is Exploring To Implement Four-day Work Week. This Research Paper Examines Of Four-day Work Week In The IT Sector And Evaluates The Extent Of Employee Awareness Regarding The Legal Framework Governing Working Hours, Over Time, And Flexible Working Arrangements. Productivity Often Increased Due To Shorter And More Focused Work Periods. The Research Focuses On The Key Indicators Such As Productivity Levels, Employee Job Satisfaction, Stress Reduction, Work-life Balance, Employee Retention. Additionally, The Study Investigates Whether Employees Possess Sufficient Knowledge About Relevant Labour Law Provisions Such As Working Hours, Overtime Entitlements Contractual Obligations And Employee Responsibilities Under Indian Labour Laws And Related Workplace Regulations. Read more Read less

IoT-ENABLED SOLAR DRIP IRRIGATION WITH ELECTROMAGNETIC WATER TREATMENT AND AUTOMATIC CLOGGING CONTROL

This Paper Proposes IoT-enabled Solar Drip Irrigation With Electromagnetic Water Treatment And Automatic Clogging Control. Due To Mineral Precipitation And Sediments, Drip Irrigation Systems Reduce Efficiency By 50% And Also Suffer Frequent Emitter C... This Paper Proposes IoT-enabled Solar Drip Irrigation With Electromagnetic Water Treatment And Automatic Clogging Control. Due To Mineral Precipitation And Sediments, Drip Irrigation Systems Reduce Efficiency By 50% And Also Suffer Frequent Emitter Clogging. Water Mineralization Modifies The Electromagnetic Fields, While Pressure Differential Sensors Are Used To Trigger Solenoid Flushing Valves To Clear Blockages. EC/TDS, Integrated Flow, Temperature, Soil Moisture, And Humidity Sensors Are Used For Precision Irrigation With Farmer Alerts. Hardware Prototype Testing Demonstrates 80-90% Flow Recovery Post Flushing And 25% Water Savings. This System Is Used To Eliminate 15-40% Energy Cost By Using Solar Power And Increase 15% Crop Yield, Which Is Suitable For Tamil Nadu’s Agricultural Conditions. Read more Read less

SEEPAGE ANALYSIS IN PARTITION WALLS & ITS CONTROL MEASURES

Seepage In Partition Walls Is A Common Issue In Buildings, Leading To Structural Deterioration, Aesthetic Damage, And Health Hazards. It Occurs Due To The Movement Of Water Through Porous Materials, Cracks, Or Construction Defects. This Paper Present... Seepage In Partition Walls Is A Common Issue In Buildings, Leading To Structural Deterioration, Aesthetic Damage, And Health Hazards. It Occurs Due To The Movement Of Water Through Porous Materials, Cracks, Or Construction Defects. This Paper Presents An Analysis Of Seepage Mechanisms In Partition Walls, Identifies Major Causes, And Discusses Effective Control And Preventive Measures. The Study Emphasizes The Importance Of Proper Design, Construction Practices, And Maintenance Strategies To Mitigate Seepage-related Problems. Read more Read less

YOLO-MineSafe: A Vision-Based Abnormal Fall Detection And Emergency Alert Framework For Isolated Mining Workers

Mining Operations Consistently Rank Among The World’s Most Hazardous Occupational Environments, With Workers Stationed In Isolated Areas Facing Undetected Fall Risks, Sudden Health Emergencies, And Life-threatening Incidents That Current Safety Syste... Mining Operations Consistently Rank Among The World’s Most Hazardous Occupational Environments, With Workers Stationed In Isolated Areas Facing Undetected Fall Risks, Sudden Health Emergencies, And Life-threatening Incidents That Current Safety Systems Cannot Address In Real-time. Existing Solutions, Such As Passive Closed-circuit Television (CCTV), Wearable Accelerometers, And Manual Supervision, Fail To Deliver Autonomous, Real-time Incident Detection Across The Expansive And Harsh Terrain Of Active Mine Sites. This Study Introduces YOLO-MineSafe, A Vision-based Fall Detection And Emergency Alert Framework Purpose-built To Close This Gap. The System Continuously Processes Surveillance Camera Videos Using A Fine-tuned YOLOv8 Deep Learning Model, Extracting Bounding Box Geometry, Posture Orientation, And Inter-frame Motion Vectors To Identify Anomalous Body Positions. A Temporal Classification Module Employing A 20-frame Confirmation Window At A 0.4 Confidence Threshold Reliably Distinguished Genuine Fall Events From Ordinary Work Postures, Such As Bending Or Crouching. Upon Confirmed Detection, Multichannel Emergency Alerts Are Dispatched Immediately: An Annotated Incident Image Via Email, An SMS To Registered Supervisors, And A Simultaneous Local Audio Alarm — All Without Human Intervention. The System Operates Effectively In Low-light And Dust-heavy Environments Through Dedicated Preprocessing, Requires No Wearable Devices, And Provides A Complete Incident Audit Trail, Representing A Substantive Advance Toward Reducing Preventable Fatalities In Isolated Mining Environments. Read more Read less

Identifying Credit Card Frauds Employing Deep Learning

With Increased Internet Usage, Online Transactions Have Been On The Rise. One Of The Most Prevalent Problems Faced Is Credit Cards Frauds. While Web Applications And Mailing Services Are Heavily Spammed, The Upsurge Of Handheld Mobile Devices Has Led... With Increased Internet Usage, Online Transactions Have Been On The Rise. One Of The Most Prevalent Problems Faced Is Credit Cards Frauds. While Web Applications And Mailing Services Are Heavily Spammed, The Upsurge Of Handheld Mobile Devices Has Led To An Outburst Of Heavy Mobile Credit Card Spamming. The Matter Is More Severe In Mobile Devices Due To Lesser Sophisticated Filtering Mechanisms In Built In Mobile Operating Systems.Recent Advancements In Electronic Commerce And Communication Systems Have Significantly Increased The Use Of Credit Cards For Both Online And Regular Transactions. However, There Has Been A Steady Rise In Fraudulent Credit Card Transactions, Costing Financial Companies Huge Losses Every Year. The Development Of Effective Fraud Detection Algorithms Is Vital In Minimizing These Losses, But It Is Challenging Because Most Credit Card Datasets Are Highly Imbalanced. This Work Proposes A Supervised Machine Learning Algorithm To Be Trained To Detect Credit Card Frauds Based On The BayesNet With Penalty Based Regularization. It Is Shown That The Proposed Approach Attains Higher Classification Accuracy Compared To Existing Work. Read more Read less

A Review On Estimating Cloud Performance Metrics Using Machine Learning And Deep Learning Models

Data Driven Cloud Computing Model Have Resulted In Unprecedented Paradigm Shifts In Cloud Application Development. Many Applications Have Found Data Driven Cloud Computing Models Indispensable Due To The Need For High Performance Computing. Performan... Data Driven Cloud Computing Model Have Resulted In Unprecedented Paradigm Shifts In Cloud Application Development. Many Applications Have Found Data Driven Cloud Computing Models Indispensable Due To The Need For High Performance Computing. Performance Prediction Is Essential For Both Cloud Service Providers And Users. Providers Rely On Accurate Predictions To Manage Resources Effectively, Prevent Over-provisioning Or Under-provisioning, And Maintain Service-level Agreements (SLAs). Users, On The Other Hand, Benefit From Performance Prediction When Selecting Cloud Services That Meet Their Application Requirements. Inadequate Performance Prediction Can Lead To Increased Operational Costs, Degraded Service Quality, And Customer Dissatisfaction. Thus, Robust Prediction Mechanisms Are Indispensable In Ensuring The Efficient Operation Of Cloud Systems. This Work Presents A Regression Learning Based Model For Performance Prediction In Cloud Environments. This Paper Presents A Review On The Contemporary Machine Learning And Deep Learning Models For Estimating Cloud Performance Metrics. Read more Read less

A Novel Dense-Swish-CNN With Bi-LSTM Framework For Image Deepfake Detection

The Rapid Advancement Of Deep Generative Models, Particularly Generative Adversarial Networks (GANs) And Diffusion-based Architectures, Has Substantially Lowered The Barrier To Producing Photorealistic Synthetic Human Faces, Collectively Referred To... The Rapid Advancement Of Deep Generative Models, Particularly Generative Adversarial Networks (GANs) And Diffusion-based Architectures, Has Substantially Lowered The Barrier To Producing Photorealistic Synthetic Human Faces, Collectively Referred To As Deepfakes. Such Media Present Critical Societal Risks Encompassing Identity Fraud, Large-scale Misinformation, And Coordinated Cybercrime. Existing Detection Approaches, Predominantly Convolutional Neural Network (CNN)-based Architectures, Demonstrate Adequate Performance On Benchmark Datasets; However, They Are Limited In Their Capacity To Jointly Model Spatial Artifact Patterns And Sequential Feature Dependencies Inherent In Manipulated Imagery. This Paper Proposes A Novel Hybrid Deep Learning Framework—the Dense-Swish Convolutional Neural Network Integrated With A Bidirectional Long Short-Term Memory (Bi-LSTM) Network—designed To Overcome These Limitations. The Proposed Architecture Leverages DenseNet121 As The Backbone For Dense Multi-scale Spatial Feature Extraction, Augmented By The Swish Activation Function To Improve Gradient Propagation And Representational Capacity. Extracted Feature Maps Are Spatially Reshaped Into Sequential Vectors And Processed By A Bi-LSTM Module That Captures Bidirectional Contextual Dependencies, Thereby Enhancing Discriminative Power Against Sophisticated Forgeries. Empirical Evaluation On A Curated Real-and-fake Image Dataset Yields A Classification Accuracy Of 99.37%, Precision Of 99.44%, Recall Of 99.31%, And F1-score Of 99.37%, Representing Consistent Improvements Over CNN-only, DenseNet Transfer Learning, And Dense-Swish-CNN Baselines. Deployment Is Realized Through A Flask-based Web Application Supporting Real-time Image Upload And Classification Inference. Read more Read less

AI -POWER BASED BLIND AND VISUALLY IMPAIRED SYSTEM FOR SMART GLASS

Visual Impairment Creates Major Challenges In Daily Life And Mobility, As Individuals Often Struggle To Detect Obstacles And Recognize People Around Them. To Address These Issues, Smart Assistive Technologies Play A Vital Role In Improving Independen... Visual Impairment Creates Major Challenges In Daily Life And Mobility, As Individuals Often Struggle To Detect Obstacles And Recognize People Around Them. To Address These Issues, Smart Assistive Technologies Play A Vital Role In Improving Independence And Safety. This Project Uses Object Detection To Identify Items Ahead, Helping Users Navigate Their Environment More Effectively. It Also Includes A Face Recognition System To Identify Both Familiar And Unfamiliar Individuals. A Deep Learning-based Convolutional Neural Network (CNN) Processes Images Captured By A Camera And Classifies Them Accurately. The Detected Information Is Then Converted Into Audio Output, Providing Voice-based Guidance To The User And Offering A More Efficient Solution Compared To Traditional Methods Like White Canes Read more Read less

AI-Based Intelligent Multi-Disease Prediction System Using Adaptive Symptom Analysis - A Comparative Study Of Supervised Learning Algorithms For Clinical Disease Classification

The Integration Of Artificial Intelligence Into Clinical Healthcare Settings Has Gained Remarkable Momentum Over The Past Decade, Offering Unprecedented Opportunities To Augment Diagnostic Capabilities And Extend Medical Reach To Underserved Populati... The Integration Of Artificial Intelligence Into Clinical Healthcare Settings Has Gained Remarkable Momentum Over The Past Decade, Offering Unprecedented Opportunities To Augment Diagnostic Capabilities And Extend Medical Reach To Underserved Populations. Among The Many Promising Applications Of This Integration, The Automated Identification Of Diseases From Patient-reported Symptom Profiles Stands Out For Its Potential To Democratize Access To Preliminary Healthcare Screening. This Paper Presents An Empirical Comparative Investigation Of Four Supervised Machine Learning Algorithms — Random Forest, K-Nearest Neighbors, Naive Bayes, And Support Vector Machine — Evaluated On The Task Of Predicting Diseases From Binary Symptom Feature Vectors. Experiments Were Conducted On A Structured Dataset Containing 4,920 Patient Records Distributed Across 41 Disease Categories And Encoded Using 131 Symptom Attributes. A Stratified 80/20 Train-test Partition Was Employed Alongside 5-fold Cross-validation To Ensure Reliable And Generalizable Performance Estimates. Among The Four Algorithms Evaluated, The Support Vector Machine Equipped With A Radial Basis Function Kernel Consistently Outperformed Its Counterparts, Attaining A Test Accuracy Of 99.19%, A Weighted F1-score Of 99.16%, And A Cross-validation Mean Of 99.27% — Results That Clearly Exceeded The 90% Performance Target Established At The Outset Of This Research. Beyond The Algorithmic Investigation, The Paper Describes The Development Of A Fully Functional Flask-based Web Application Named MediAI, Which Embeds The Trained SVM Model Within An Adaptive Symptom Collection Interface Capable Of Delivering Real-time Differential Diagnoses, Confidence-calibrated Predictions, And Actionable Clinical Guidance. These Findings Collectively Affirm That Classical Supervised Learning, When Thoughtfully Applied To Well-structured Clinical Data, Can Serve As A Reliable Foundation For Accessible And Scalable Disease Screening Systems. Read more Read less

BLOCKCHAIN-BASED MULTI-USER DYNAMIC VERIFIABLE SEARCHABLE ENCRYPTION FOR SECURE DATA STORAGE AND QUERY ON MALICIOUS CLOUD SERVER

The Increasing Elderly Population Has Led To A Growing Demand For Intelligent Healthcare Monitoring Systems. Falls Are One Of The Major Causes Of Injury And Mortality Among Elderly Individuals. This Paper Proposes An IoT-based Remote Monitoring Syste... The Increasing Elderly Population Has Led To A Growing Demand For Intelligent Healthcare Monitoring Systems. Falls Are One Of The Major Causes Of Injury And Mortality Among Elderly Individuals. This Paper Proposes An IoT-based Remote Monitoring System Using Six-axis Acceleration Sensors To Detect Falls In Real Time. The System Integrates Edge Computing To Process Sensor Data Locally, Reducing Latency And Improving Response Time. By Combining Accelerometer And Gyroscope Data, The System Accurately Distinguishes Between Normal Activities And Fall Events. Alerts Are Instantly Sent To Caregivers Through IoT Communication Modules. The Proposed System Ensures Improved Safety, Reduced False Alarms, And Enhanced Real-time Monitoring For Elderly Individuals. Read more Read less

A Study On Cost Analysis And Its Importance In Business Decision Making

This Study Examines Cost Analysis And Its Importance In Business Decision Making.Cost Analysis Helps Businesses Identify And Understand Different Types Of Costs.The System Enables Managers To Manage Their Expenses By Using Its Functions For Planning... This Study Examines Cost Analysis And Its Importance In Business Decision Making.Cost Analysis Helps Businesses Identify And Understand Different Types Of Costs.The System Enables Managers To Manage Their Expenses By Using Its Functions For Planning And Expense Control And Cost Reduction. The Study Focuses On How Cost Information Influences Managerial Decisions. The Study Demonstrates That Better Understanding Of Fixed And Variable Costs Leads To Improved Development Of Pricing Models. The Study Provides Information About How Organizations Use Financial Resources To Develop Their Strategic Plans And Operational Execution. The Research Demonstrates That Marginal Cost Information Helps Organizations Make Decisions About Their Short-term Business Operations. Researchers Collected Data By Using A Structured Questionnaire To Obtain Information From Participants. The Results Demonstrate That Cost Analysis Provides Businesses With Essential Insights That Enhance Their Operational Effectiveness. The Study Establishes That Cost Analysis Serves As A Fundamental Requirement For Organizations To Achieve Successful Decision Making Processes. Read more Read less

AM Biofertilizer And Sustainable Resource Management For Arid Land Cultivation Practice

Arbuscular Mycorrhiza (AM) Plays A Crucial Role To Combat The Water Stress Condition In Dry Land For The Host Plants. More Than 80% Of The Angiosperm Plants Are AM Infected Symbiotically. So, The Use Is Beneficial And Application Is Very Crucial As B... Arbuscular Mycorrhiza (AM) Plays A Crucial Role To Combat The Water Stress Condition In Dry Land For The Host Plants. More Than 80% Of The Angiosperm Plants Are AM Infected Symbiotically. So, The Use Is Beneficial And Application Is Very Crucial As Biofertlizers. In This Communication Broad Aspects Of AM Use Pattern And Its Global Perspective Have Been Presented. Result Revealed That Natural Forest With AM Spore Density Affect Positively On Geophyte Yield. Read more Read less

A STUDY ON THE IMPACT OF HYBRID REMOTE WORK ON CAREER PROGRESSION AND WELL-BEING OF EMPLOYEES WORKING FROM REGIONAL HOMETOWNS

This Study Examines The Impact Of Hybrid Remote Work On Career Progression And Well-being Of Employees Working From Regional Hometowns. With The Increasing Adoption Of Flexible Work Models, Employees Are No Longer Restricted To Metropolitan Workplace... This Study Examines The Impact Of Hybrid Remote Work On Career Progression And Well-being Of Employees Working From Regional Hometowns. With The Increasing Adoption Of Flexible Work Models, Employees Are No Longer Restricted To Metropolitan Workplaces. The Study Aims To Understand Employee Motivations, Career Growth Opportunities, And Work-life Balance Under Hybrid Work Conditions. Primary Data Was Collected From 84 Respondents In The IT Sector Using Structured Questionnaires. Statistical Tools Such As Percentage Analysis, Descriptive Statistics, Correlation, Chi-square Test, And Friedman Test Were Used. The Findings Reveal That Hybrid Work Enhances Flexibility, Reduces Living Costs, And Improves Well-being. However, Challenges Such As Limited Professional Visibility, Communication Barriers, And Infrastructure Issues Affect Career Progression. The Study Concludes That Organizations Must Develop Structured Policies To Ensure Equal Career Opportunities And Support Employee Well-being. Read more Read less

Online Fraud Transaction Detection Using XGBoost, PCA, And CNN1D

The Online Payment Platforms And Financial Services Are Growing Fast. This Means That Online Fraud Transactions Are Also Increasing. Old Methods Of Detecting Fraud Do Not Work Well. They Cannot Find The Changing Fraud Schemes. We Propose A System Tha... The Online Payment Platforms And Financial Services Are Growing Fast. This Means That Online Fraud Transactions Are Also Increasing. Old Methods Of Detecting Fraud Do Not Work Well. They Cannot Find The Changing Fraud Schemes. We Propose A System That Uses Machine Learning To Detect Online Fraudulenttransactions. This System Uses Extreme Gradient Boosting, Principal Component Analysis, And One-dimensional Convolutional Neural Network. Our System Is Designed To Detect Online Fraudulent Transactions Efficiently. It Uses A Combination Of Machine Learning Models To Identifypatterns In Transaction Data. The System Also Uses Data Balancing Methods To Address The Class Imbalance Problem. We Tested Our System. It Performed Better Than Traditional Methods. The Results Show That Our System Is Better At Detecting Fraudulent Transactions. It Also Reduces The Number Of Positive Results. Our System A Solution For Securing Online Financial Transactions. Read more Read less

Electricity Generation Using Wind Power

This Expanded Report Presents A More Technical And Calculation-oriented Study Of A Small PVC Windmill Electricity Generation System. The Report Preserves The Academic Chapter Structure Of The Sample Report Supplied By The User While Changing The Proj... This Expanded Report Presents A More Technical And Calculation-oriented Study Of A Small PVC Windmill Electricity Generation System. The Report Preserves The Academic Chapter Structure Of The Sample Report Supplied By The User While Changing The Project Topic From Cycle-based Generation To Wind-based Generation. The Work Is Anchored To The User-supplied YouTube Project Topic, Namely A Homemade PVC Windmill For Electricity Generation, And Is Further Developed Into A Formal Engineering-style Document Through The Addition Of Assumed Design Specifications, Analytical Calculations, Electrical Architecture, And Multiple Diagrams. The Study Explains How A PVC Rotor Extracts Energy From Wind, Converts It Into Shaft Torque, Drives A 24VDC Generator, Conditions The Variable Output Through Rectification And Regulation, And Finally Delivers Power To A Battery Or Low-voltage DC Load. A Detailed Hardware Chapter Is Included To Document The Rotor, Hub, Shaft, Bearings, Generator, Rectifier, Capacitor, Controller, Battery, And Measuring Instruments. The Report Also Introduces Standard Wind-energy Equations Such As Rotor Swept Area, Available Wind Power, Power Coefficient, Electrical Output, Tip-speed Ratio, Rotor-speed Estimation, And Battery Charging Time. These Calculations Are Clearly Marked As Educational Engineering Estimates Because The Source Video Itself Does Not Publish A Full Bill Of Materials Or Measured Laboratory Data. The Resulting Document Is Suitable For Project Submission, Viva Preparation, Or Further Customization Into Institution- Specific Format. It Is Especially Useful For Students Who Need A Longer Project Report With More Technical Depth, Better Organization, And Stronger Numerical Content Than A Simple Overview Report. Read more Read less

Determinants Of The Attitude–Behavior Gap In Sustainable Consumption An Empirical Analysis Of Youth Consumers

Sustainable Consumption Has Gained Significant Importance In Recent Years, Particularly Among Youth Who Are Increasingly Aware Of Environmental Issues. However, A Noticeable Gap Exists Between Positive Attitudes Toward Sustainability And Actual Purch... Sustainable Consumption Has Gained Significant Importance In Recent Years, Particularly Among Youth Who Are Increasingly Aware Of Environmental Issues. However, A Noticeable Gap Exists Between Positive Attitudes Toward Sustainability And Actual Purchasing Behavior. This Study Examines The Attitude–behavior Gap Among Young Consumers Using Primary Data Collected From 50 Respondents. The Findings Reveal High Awareness And Favorable Attitudes Toward Sustainable Products, But Relatively Lower Levels Of Actual Purchasing Behavior. Key Barriers Identified Include High Prices, Limited Availability, And Lack Of Trust In Sustainability Claims. The Study Highlights The Need For Improved Accessibility, Affordability, And Transparency To Bridge This Gap. Read more Read less

Hybrid Deep Learning Model For Early Fault Detection In Energy-Intensive Tablet Press Equipment: MLP–1D CNN Fusion For EMIS Applications

Fault Detection In Pharmaceutical Tablet Press Equipment Is Crucial For Ensuring Product Quality, Minimizing Downtime, And Reducing Energy Consumption In Energy-intensive Manufacturing Environments. This Study Presents A Hybrid Deep Learning Model, N... Fault Detection In Pharmaceutical Tablet Press Equipment Is Crucial For Ensuring Product Quality, Minimizing Downtime, And Reducing Energy Consumption In Energy-intensive Manufacturing Environments. This Study Presents A Hybrid Deep Learning Model, Namely MLP–1D CNN FaultNet, Which Integrates A Multilayer Perceptron (MLP) And A One-dimensional Convolutional Neural Network (1D CNN). The Architecture Employs Parallel Branches To Capture Both Global Statistical Dependencies And Localized Feature Patterns. The MLP Branch Models Global Feature Interactions, While The 1D CNN Branch Extracts Spatial Correlations Through Convolutional Operations. The Learned Representations Are Fused In A Dedicated Layer And Further Refined Using Dense Layers With Dropout And Batch Normalization To Improve Generalization. The Final Classification Layer Performs Fault Detection Effectively. Experimental Results Indicate That The Hybrid Model Outperforms Standalone MLP And CNN Models In Terms Of Accuracy, Precision, Recall, And F1-score. Therefore, The Architecture Is Suitable For Real-time Monitoring, Predictive Maintenance, And Energy-aware Fault Management In Pharmaceutical Manufacturing Systems. Read more Read less

Automated Football Detection, Tracking, And Object Detection, And Homography

Modern Football Has Transitioned Into A Data-centric Era, Where Tactical Efficiency Is Often Measured Through Granular Metrics Rather Than Just Match Outcomes. This Project Addresses The Gap Between Raw Broadcast Footage And Structured Analytical Dat... Modern Football Has Transitioned Into A Data-centric Era, Where Tactical Efficiency Is Often Measured Through Granular Metrics Rather Than Just Match Outcomes. This Project Addresses The Gap Between Raw Broadcast Footage And Structured Analytical Data By Constructing A Custom Computer Vision Pipeline. Instead Of Relying On Expensive, Proprietary Tracking Systems Used By Elite Clubs, We Developed A Modular System Capable Of Extracting Player Trajectories, Team Formations, And Physical Performance Indicators From Standard Single-camera Video Feeds. Our Implementation Integrates YOLOv8 For Robust Object Detection With The SORT Algorithm For Real-time Tracking, Enhanced By A Custom Homography-based Camera Motion Compensation Module. A Key Focus Of Our Work Was Mathematically Decoupling The Camera’s Panning And Zooming Movements From The Actual Player Velocity, A Common Source Of Error In Amateur Ana- Lytics Projects. By Mapping Pixel Coordinates To A Real-world Pitch Model, We Successfully Derived Actionable Insights Such As Heatmaps And Sprint Profiles. This Paper Detailed The Specific Engineering Challenges Encountered—from Handling Occlusion In Crowded Penalty Boxes To Calibrating Color Thresholds For Jersey Segmentation—and Presents A Scalable, Open-source Approach To Democratizing Sports Analytics. Read more Read less

Voice Based Secure Transaction System With Deepfake Detection

The Development And Growth Of Digital Payment Systems Have Created A Need For Secure Authentication Mechanisms. The Use Of Passwords, PINs And One-time Passwords Is No Longer Secure Against Cyber Attacks And Identity Theft. The Paper Suggests A Voice... The Development And Growth Of Digital Payment Systems Have Created A Need For Secure Authentication Mechanisms. The Use Of Passwords, PINs And One-time Passwords Is No Longer Secure Against Cyber Attacks And Identity Theft. The Paper Suggests A Voice Based Secure Transaction System With Deep Fake Detection. The Suggested System Will Use Machine Learning Algorithms For Speaker Verification And Deep Fake Detection. The MFCC Method Will Be Used For Feature Extraction. The User’s Voice Command Will Be Converted To Text By A Speech Recognition System. The Text Will Be Compared To The Saved Voice Samples For Verification. The Deep Fake Detection Module Will Check Whether The Voice Is Genuine Or Not. If The Voice Is Genuine, The Transaction Will Proceed Otherwise, The Transaction Will Be Denied. The Suggested System Is Efficient And Secure For Digital Payment Systems. Read more Read less

VIDEO CAPTIONING AND EMOTION RECOGNITION USING CNN+LSTM

With The Rapid Growth Of Digital Video Content, Particularly Across Social Media Platforms, Short And Engaging Videos Have Become Increasingly Dominant In Capturing User Attention. Video Captioning Plays A Critical Role In Addressing This Trend By Au... With The Rapid Growth Of Digital Video Content, Particularly Across Social Media Platforms, Short And Engaging Videos Have Become Increasingly Dominant In Capturing User Attention. Video Captioning Plays A Critical Role In Addressing This Trend By Automatically Generating Descriptive Textual Representations Of Video Content, Thereby Improving Accessibility And Enhancing User Engagement. The Process Of Video Captioning Involves Two Primary Stages: Feature Extraction And Caption Generation. In This Work, Pre-trained Convolutional Neural Networks (CNNs), Such As InceptionV3 And VGG16, Are Employed To Extract High-level Visual Features From Video Frames. These Extracted Features Are Subsequently Provided As Input To A Long Short-Term Memory (LSTM) Network, Which Generates Contextually Coherent Captions.The Incorporation Of LSTM Networks In Conjunction With Word Embeddings Facilitates The Generation Of Semantically Meaningful Captions While Enabling Effective Emotion Classification. This Integrated Framework Significantly Enhances The Overall Understanding Of Video Content. Overall, This Work Presents A Comprehensive And Efficient Solution For Intelligent Video Interpretation By Integrating Visual Feature Extraction With Contextual And Emotional Analysis, Thereby Advancing The Capabilities Of Automated Multimedia Understanding Systems. Read more Read less

A Study On Algorithm-Based Pricing In Digital Platforms

Algorithm-based Pricing Is Increasingly Used By Digital Platforms To Set Prices Dynamically Based On Data, Demand, And User Behavior. This Study Examines How Such Pricing Mechanisms Operate And Their Impact On Consumers, Market Competition, And Fairn... Algorithm-based Pricing Is Increasingly Used By Digital Platforms To Set Prices Dynamically Based On Data, Demand, And User Behavior. This Study Examines How Such Pricing Mechanisms Operate And Their Impact On Consumers, Market Competition, And Fairness. The Research Adopts Both Doctrinal And Non-doctrinal Methods To Analyze Legal Principles, Regulatory Concerns, And Real-world Practices. Primary Data Is Collected Through A Structured Google Form Survey, While Secondary Sources Include Statutes, Case Laws, And Scholarly Articles. The Study Aims To Identify Transparency Issues, Potential Discrimination, And Regulatory Gaps In Algorithm-driven Pricing Systems. The Findings Seek To Contribute To Better Understanding And Policy Development In The Digital Economy. Read more Read less

An Autonomous Orchestration Platform For Intelligent Campus Management.

The Rapid Growth Of Digital Education Systems Has Created A Need For Efficient And Intelligent Campus Management Solutions. Traditional Campus Systems Such As Help Desks, Attendance Tracking, Fee Management, And Result Processing Often Operate Indepe... The Rapid Growth Of Digital Education Systems Has Created A Need For Efficient And Intelligent Campus Management Solutions. Traditional Campus Systems Such As Help Desks, Attendance Tracking, Fee Management, And Result Processing Often Operate Independently, Leading To Inefficiencies, Delays, And Lack Of Real-time Communication. To Address These Challenges, This Paper Proposes An Autonomous Orchestration Platform For Intelligent Campus Management. The System Utilizes A Multi-agent Architecture Where Different Agents Handle Specific Tasks Such As Attendance Monitoring, Fee Management, Result Processing, And Query Handling. A Central Coordinator Agent Orchestrates And Manages Communication Between All Modules, Ensuring Smooth Workflow Execution. The Integration Of Artificial Intelligence Enables Intelligent Decision-making And Personalized Interaction. Additionally, Workflow Automation Tools Are Used To Trigger Real-time Notifications And Alerts, Improving Efficiency And Reducing Manual Effort. The Proposed System Provides A Centralized, Scalable, And Automated Solution For Modern Campus Management. Read more Read less

Intelligent Network Traffic Prediction Using Graph Neural Networks

The Widespread Deployment Of Wireless Communication Technology Has Made The Wi-Fi Network More Susceptible To Sophisticated Cyber-attacks, Which Include De-authentication Attacks, Rogue Access Points, Traffic Flooding, And Protocol Misuse. The Limita... The Widespread Deployment Of Wireless Communication Technology Has Made The Wi-Fi Network More Susceptible To Sophisticated Cyber-attacks, Which Include De-authentication Attacks, Rogue Access Points, Traffic Flooding, And Protocol Misuse. The Limitation Of The Traditional Intrusion Detection System In Detecting Unknown Attacks In Real-time Has Led To The Development Of This Paper, Which Presents An Intelligent Real-time Detection System For Wi-Fi Network Attacks Using Machine Learning And Graph Neural Networks (GNNs) Technology. The Proposed System Monitors The Wi-Fi Network Traffic In Real-time And Extracts Essential Features From The IEEE 802.11 Protocol Frames, Which Include Packet Rates, Signal Strength, Protocol Behaviour, And Time-based Features. The Proposed System Uses The Extracted Features To Classify The Network Behaviour As Normal Or Abnormal Using The Trained Machine Learning Models. The Proposed System Effectively Uses The GNN Technology To Detect The Complex Relationships Between The Network Entities, Which Makes It More Efficient In Detecting Network Intrusions. The Proposed System Has Been Tested And Found Effective In Terms Of Accuracy, Reducing False Positives, And Ensuring Real-time Detection Of Network Intrusions, Which Makes It More Suitable For Modern Enterprise And Public Network Environments. Read more Read less

An Intelligent System For Identifying Fake Reviews In E-commerce Websites

The Rapid Growth Of E-commerce Platforms Has Increased The Influence Of Online Customer Reviews On Purchasing Decisions. However, The Presence Of Fake Or Deceptive Reviews Has Become A Major Challenge, Misleading Customers And Affecting The Credibili... The Rapid Growth Of E-commerce Platforms Has Increased The Influence Of Online Customer Reviews On Purchasing Decisions. However, The Presence Of Fake Or Deceptive Reviews Has Become A Major Challenge, Misleading Customers And Affecting The Credibility Of Online Marketplaces. Detecting Such Fraudulent Reviews Is Essential To Ensure Trust And Transparency In Digital Commerce. This Project Proposes A Machine Learning-based Fake Review Detection System Integrated Into A Web-based E-commerce Environment. The System Allows Users To Browse Products And Submit Reviews, Which Are Automatically Analyzed Using Natural Language Processing (NLP) Techniques And Classified As Genuine Or Fake Using Decision Tree And XGBoost Algorithms. The Reviews Are Preprocessed And Converted Into Numerical Features Using The TF-IDF Technique. The Models Are Evaluated Using Performance Metrics Such As Accuracy, Precision, Recall, And F1-Score. The Classified Results Are Stored In A Database, And Only Genuine Reviews Are Considered For Product Ratings. An Admin Dashboardprovides Insights Into Fake Review Statistics And Model Performance. By Integrating Machine Learning With A Real-time Web Application, The Proposed System Improves The Reliability Of Online Reviews And Enhances User Trust In E-commerce Platforms. Read more Read less

AEROVOLT

AeroVolt Is A Compact And Efficient Renewable Energy System Designed To Generate Electricity Using Wind Power Through A Vertical Axis Wind Turbine (VAWT). The Main Objective Of This Project Is To Utilize Low-speed Wind Energy Available In Urban And R... AeroVolt Is A Compact And Efficient Renewable Energy System Designed To Generate Electricity Using Wind Power Through A Vertical Axis Wind Turbine (VAWT). The Main Objective Of This Project Is To Utilize Low-speed Wind Energy Available In Urban And Rural Areas And Convert It Into Useful Electrical Energy In An Eco-friendly Manner. The System Consists Of Vertical Blades Mounted Around A Central Shaft, Which Rotate When Wind Flows From Any Direction. This Rotational Motion Is Transferred To A Generator, Where Mechanical Energy Is Converted Into Electrical Energy. Unlike Traditional Horizontal Wind Turbines, AeroVolt Does Not Require Alignment With Wind Direction, Making It More Suitable For Locations With Irregular Wind Patterns. The Design Is Simple, Cost-effective, And Requires Low Maintenance, Making It Ideal For Small-scale Applications Such As Street Lighting, Household Power Generation, And Educational Purposes. The AeroVolt System Contributes To Reducing Dependence On Fossil Fuels And Supports Sustainable Development By Promoting Clean And Green Energy. In Conclusion, AeroVolt Demonstrates An Innovative Approach To Harness Wind Energy Efficiently In A Compact Form, Making It A Promising Solution For Future Renewable Energy Needs. Read more Read less

AI DRIVEN SMART SUPPLY CHAIN MANAGEMENT SYSTEM

Demand Forecasting Plays A Pivotal Role In Modern Supply Chain Management, Directly Influencing Inventory Planning, Logistics, And Overall Business Decision-making. Traditional Forecasting Systems Often Rely On Static Machine Learning Models That Gra... Demand Forecasting Plays A Pivotal Role In Modern Supply Chain Management, Directly Influencing Inventory Planning, Logistics, And Overall Business Decision-making. Traditional Forecasting Systems Often Rely On Static Machine Learning Models That Gradually Lose Accuracy As Market Conditions Evolve, Leading To Inefficiencies And Poor Resource Utilization. To Address These Limitations, This Paper Proposes An AIOps-driven Demand Forecasting System Built On The MERN Stack. The System Integrates Machine Learning Models Such As LSTM And Prophet With An AIOps Monitoring Layer That Continuously Tracks Performance, Detects Data Drift, And Triggers Automatic Retraining When Necessary. By Combining Scalable Web Technologies With Adaptive AI, The Proposed Solution Delivers A Self- Improving, Real-time Forecasting Platform That Enhances Supply Chain Resilience, Reduces Stockouts, And Supports Intelligent Business Decisions. Read more Read less

ENHANCED NET BANKING SECURITY USING ILLUSION PIN, FACIAL BIOMETRICS, AND COLLABORATIVE TRANSACTION

This Paper Presents An Enhanced Net Banking Security Framework Designed To Overcome The Limitations Of Traditional Authentication Mechanisms Such As Passwords And PINs, Which Are Increasingly Vulnerable To Cyber Threats Including Phishing, Shoulder S... This Paper Presents An Enhanced Net Banking Security Framework Designed To Overcome The Limitations Of Traditional Authentication Mechanisms Such As Passwords And PINs, Which Are Increasingly Vulnerable To Cyber Threats Including Phishing, Shoulder Surfing, And Brute-force Attacks. This System Introduces A Multi-layered Authentication Approach Combining Illusion PIN Techniques With Real-time Facial Biometric Verification. Blockchain Technology Ensures Secure, Transparent, And Tamper-proof Storage Of Transaction Records. A Joint Account Multi-Party Authorization (MPA) Module Enables Collaborative Transaction Approvals, Requiring Consent From Multiple Authorized Users Before Executing Sensitive Operations. By Integrating These Technologies, The Proposed System Delivers A Comprehensive, Secure, And User-friendly Net Banking Environment That Minimizes Fraud And Ensures Robust Protection Of Financial Data. Read more Read less

Blockchain-Based Multi-User Dynamic Verifiable Searchable Encryption For Secure Data Storage And Query On Malicious Cloud Server

Cloud Computing Has Become The Dominant Platform For Scalable Data Storage And Distributed Services. Despite Its Advantages, Outsourcing Sensitive Data To Untrusted Cloud Servers Introduces Confidentiality And Integrity Risks. Dynamic Searchable Symm... Cloud Computing Has Become The Dominant Platform For Scalable Data Storage And Distributed Services. Despite Its Advantages, Outsourcing Sensitive Data To Untrusted Cloud Servers Introduces Confidentiality And Integrity Risks. Dynamic Searchable Symmetric Encryption (DSSE) Enables Users To Search Encrypted Data Without Revealing Plaintext Information, Yet Most Existing Schemes Assume Single-user Environments Or Semi-honest Servers. This Paper Presents A Blockchain- Based Multi-user Dynamic Verifiable Searchable Encryption Framework Designed For Malicious Cloud Settings. The Proposed System Stores Lightweight Index Metadata On A Blockchain To Ensure Tamper- Proof Verification While Preserving Storage Efficiency. A Cuckoo Filter- Based Index Combined With A Merkle Hash Tree Enables Efficient And Verifiable Search Operations. Secure Multi-user Key Management Is Achieved Using Diffie– Hellman Key Exchange, Eliminating Reliance On Centralized Trust Authorities. Read more Read less

Colorectal Cancer Detection Using Pre-Trained Ensemble Algorithms

Colorectal Cancer Is The Second Leading Cause Of Cancer-related Deaths Globally And The Most Common Cancer After Lung And Breast Cancer. If Caught Early, Many Lives Can Be Saved. On The Other Hand, Conventional Diagnosis Requires Doctors To Manually... Colorectal Cancer Is The Second Leading Cause Of Cancer-related Deaths Globally And The Most Common Cancer After Lung And Breast Cancer. If Caught Early, Many Lives Can Be Saved. On The Other Hand, Conventional Diagnosis Requires Doctors To Manually Analyze Histopathological Images, Which Can Be A Lengthy Process, Expensive, And Sometimes Even Lead To Human Mistakes. This Creates A Strong Need For Smarter And More Reliable Systems That Can Support Medical Professionals In Their Work. This Paper Presents A Colorectal Cancer Detection System Built On Pre-trained Deep Learning Models. Our System Employs Popular Architectures Such As Inspection V3, Resnet50, And EfficientNetB0 That Are Not Only Capable Of Identifying Key Features In Medical Images But Also Help In Enhancing The Accuracy Of The Predictions. The Model Was Built And Evaluated Using The LC25000 Dataset, Which Comprises Histopathological Images Of Both Cancerous And Non-cancerous Tissues. We Do Not Need To Depend On A Single Model But Instead Use An Ensemble Approach, Where The Strengths Of Multiple Models Are Combined Into One. This Helps Us Achieve Better Results In Terms Of Accuracy, Precision, Recall, And F1-score Compared To Traditional Methods Such As Random Forest And Naive Bayes. Our Findings Also Indicate That The System Can Detect Cancer Faster And More Effectively. Overall, This Approach Provides A Simple, Reliable, And Cost-effective Solution To Assist Doctors In Diagnosing Colorectal Cancer. By Supporting Early Detection, It Helps Improve Treatment Outcomes And Gives Patients A Better Chance Of Survival. Read more Read less

AI Dark Pattern Detection System For Fair Web And App UX

The Increasing Prevalence Of Deceptive User Interface Designs, Commonly Referred To As Dark Patterns, Poses Significant Challenges To User Autonomy And Transparency In Digital Environments. These Patterns Manipulate Users Into Making Unintended Decis... The Increasing Prevalence Of Deceptive User Interface Designs, Commonly Referred To As Dark Patterns, Poses Significant Challenges To User Autonomy And Transparency In Digital Environments. These Patterns Manipulate Users Into Making Unintended Decisions, Such As Accepting Unnecessary Permissions Or Engaging With Misleading Offers. Traditional Detection Approaches Are Often Manual, Reactive, Or Limited In Scalability, Making Them Ineffective For Real-time User Protection. To Address These Limitations, This Project Proposes An Automated Dark Pattern Detection System Implemented As A Chrome Browser Extension Using Manifest V3 Architecture. The System Leverages A Rule-based Detection Mechanism Grounded In The FoSIP Framework To Identify Multiple Categories Of Manipulative Design, Including Social Engineering, Forced Actions, Interface Interference, Fake Discounts, And Persistent Elements. By Integrating Real-time DOM Analysis With The MutationObserver API, The Extension Dynamically Detects And Highlights Suspicious Elements Without Requiring User Interaction. Furthermore, The System Incorporates A Fairness Scoring Model That Quantifies The Ethical Quality Of Web Pages Based On Detected Patterns And Their Severity Levels. The Modular Architecture, Built Using Lightweight Web Technologies Such As JavaScript, HTML, And CSS, Ensures Scalability And Extensibility, Enabling Future Integration With AI-based Contextual Analysis. The Proposed Solution Provides A Proactive, User-centric Approach To Enhancing Transparency In Web Interactions, Promoting Ethical Design Practices, And Empowering Users To Make Informed Decisions While Browsing. Read more Read less

DESIGN AND IMPLEMENTATION OF A SOLAR - POWERED SMART RAIN SHIELD SYSTEM FOR HOUSEHOLD CLOTHES WITH GSM - BASED SMS AND IOT MONITORING

The Increasing Adoption Of Smart Home Technologies Has Created A Demand For Automated And Energy-efficient Household Solutions. Traditional Clothes Drying Methods Are Vulnerable To Unexpected Rainfall, Leading To Inconvenience And Repeated Drying Eff... The Increasing Adoption Of Smart Home Technologies Has Created A Demand For Automated And Energy-efficient Household Solutions. Traditional Clothes Drying Methods Are Vulnerable To Unexpected Rainfall, Leading To Inconvenience And Repeated Drying Efforts. Existing Systems Provide Rain Detection And Mechanical Covering Mechanisms But Lack SMS Notification And Remote Monitoring, Limiting User Awareness. The Proposed System Presents The Design And Implementation Of A Solar-Powered Smart Rain Shield System Integrated With GSM-based SMS Alerts And IoT Monitoring. A Rain Sensor Detects Precipitation And Sends Signals To The ESP32 Microcontroller, Which Controls A Servo Motor To Automatically Deploy A Protective Shield. Unlike Existing Models, The System Sends Real-time SMS Notifications Through A GSM Module And Enables Remote Status Checking Via A Web-based IoT Platform. The System Operates Using Solar Energy With Battery Backup, Ensuring Sustainable And Uninterrupted Performance. By Integrating Renewable Energy, Embedded Control, And Wireless Communication, The Proposed Solution Enhances Safety, Improves User Convenience, And Reduces Manual Intervention. Read more Read less

FITKIT: An AI-Powered Fitness Coach with Real-Time Posture Correction and Personalized Dietary Advisory

The FitKit Application Is An Innovative And Comprehensive Digital Fitness Solution Designed To Streamline Personal Health Tracking And Ensure Safe Workout Practices Through Artificial Intelligence. Traditional Fitness Applications Primarily Focus On... The FitKit Application Is An Innovative And Comprehensive Digital Fitness Solution Designed To Streamline Personal Health Tracking And Ensure Safe Workout Practices Through Artificial Intelligence. Traditional Fitness Applications Primarily Focus On Manual Logging And Generic Video Tutorials. FitKit Revolutionizes This By Integrating Real-time Computer Vision To Provide Live Posture Analysis And Automated Exercise Logging. Built Using Python And The Streamlit Framework, The System Leverages Google's MediaPipe Pose Detection And OpenCV To Monitor User Movements During Exercises Such As Planks, Wall Sits, And Squats. The Application Features A Robust SQLite Backend For Secure User Data Management, Progress Tracking, And Exercise History. Additionally, The System Provides Real-time Voice Feedback Using Pyttsx3 To Correct User Form Instantly, Reducing The Risk Of Injury. FitKit Also Incorporates A Personalized Dietary Advisory Module That Tailors Nutrition And Workout Intensity Based On The User's BMI, Fitness Goals, And Dietary Preferences. This Project Aims To Replace Unstructured Workout Routines With An Intelligent, Automated Coach That Enhances User Motivation, Ensures Proper Form, And Provides Actionable Health Insights. Read more Read less

AI-BASED ENERGY MANAGEMENT SYSTEM

The Rapid Growth Of Energy Demand, Increasing Electricity Costs, And The Need For Sustainable Power Utilization Present Significant Challenges To Traditional Energy Management Practices. This Paper Presents An AI-Based Smart Energy Management System... The Rapid Growth Of Energy Demand, Increasing Electricity Costs, And The Need For Sustainable Power Utilization Present Significant Challenges To Traditional Energy Management Practices. This Paper Presents An AI-Based Smart Energy Management System That Leverages Artificial Intelligence And Machine Learning Techniques To Enable Intelligent Monitoring, Forecasting, And Optimization Of Energy Consumption In Residential And Industrial Environments. The Proposed System Integrates Real-time Sensor Data, Historical Energy Usage Patterns, And Environmental Parameters To Analyze Consumption Behavior And Predict Future Energy Demand With High Accuracy The Architecture Incorporates A Multi-layered Intelligence Framework Offering Four Functional Modules. Module 1 Performs Real-time Energy Monitoring And Anomaly Detection Using Data-driven Analytics. Module 2 Applies Machine Learning–based Load Forecasting To Predict Peak Demand And Consumption Trends. Module 3 Enables Automated Energy Optimization Through AI-driven Decision-making, Dynamically Scheduling Loads To Minimize Energy Wastage And Operational Costs. Module 4 Provides Adaptive Control And User-centric Insights Through A Smart Dashboard, Ensuring Transparency And Actionable Recommendations. The System Architecture Comprises Three Integrated Components: An IoT-enabled Data Acquisition Layer For Continuous Energy Sensing, An AI-powered Analytics Engine For Prediction And Optimization, And A Cloud-based Management Platform For Visualization And Control. A Key Feature Of The Proposed Solution Is Its Autonomous Learning Capability, Which Continuously Refines Energy Optimization Strategies Based On Evolving Usage Patterns And Feedback. Experimental Evaluation Demonstrates That The System Effectively Reduces Peak Energy Consumption, Improves Energy Efficiency, And Enhances Decision-making Without Requiring Significant Infrastructure Modifications. The Results Confirm That AI-driven Smart Energy Management Is A Scalable, Cost-effective, And Sustainable Approach For Modern Power Systems, Offering A Practical Pathway Toward Intelligent And Energy-efficient Ecosystems Read more Read less

An AI-Based Intelligent Fashion System For Virtual Try-On And Explainable Personalized Recommendation

In This Paper, We Introduce A New Virtual Try-On System With The Addition Of An Explainable Suitability Analysis Framework. Our Virtual Try-On System Allows Users To Virtually Try On Clothes With Images Using A Seven-stage Pipeline. Our Approach Diff... In This Paper, We Introduce A New Virtual Try-On System With The Addition Of An Explainable Suitability Analysis Framework. Our Virtual Try-On System Allows Users To Virtually Try On Clothes With Images Using A Seven-stage Pipeline. Our Approach Differs From Existing Virtual Try-On Systems In Its Use Of A Neuro-symbolic Approach To Analyze The Suitability Of An Outfit Based On Different Occasions. We Have Used Visual Features Like Color, Texture, And Silhouette To Analyze The Suitability Of An Outfit With The Help Of A Rule-based Engine. Our Approach Combines Image Generation With Intelligent Decision- Making. Read more Read less

A Study On Exploring Laws Governing E-Commerce And Fraud

E-Commerce Has Significantly Transformed As Global Commerce. As More People Shop Online, E-commerce Has Become A Huge Target For Scammers. While It’s Convenient To Buy Things With A Click, It Has Also Opened The Door To Problems Like Identity Theft,... E-Commerce Has Significantly Transformed As Global Commerce. As More People Shop Online, E-commerce Has Become A Huge Target For Scammers. While It’s Convenient To Buy Things With A Click, It Has Also Opened The Door To Problems Like Identity Theft, Fake Websites, And Credit Card Fraud. This Paper Looks At How We Can Make Online Shopping Safer By Focusing On Three Main Areas: Better Laws, Smarter Technology, And More Aware Shoppers.In This Research, Current Laws Like The Consumer Protection Act And The Newest 2025 Guidelines Were Looked Into. Meanwhile The Laws Are Getting Stronger, There Is Still A Big Gap Because Many Shoppers Don’t Know How To Protect Themselves. This Paper Explored How Technology Like AI And Simple Steps Like Two-factor Authentication Can Act As A Shield Against Hackers And Identity Thieves Read more Read less

Fruits Freshness Detection Quality And Nutritional Analysis Using Machine Learning

In The Context Of The Food Processing Industry In The Present Day, It Is Evident That The Freshness And Quality Of Fruits Play A Major Role In Maintaining The Health And Wellness Of Consumers, Reducing Food Product Waste, And Maximizing The Efficienc... In The Context Of The Food Processing Industry In The Present Day, It Is Evident That The Freshness And Quality Of Fruits Play A Major Role In Maintaining The Health And Wellness Of Consumers, Reducing Food Product Waste, And Maximizing The Efficiency Of The Food Supply Chain. Generally Speaking, It Is Evident That Traditional Methods Of Quality Evaluation Of Fruits Are Mostly Dependent On Manual Inspection Techniques. However, It Is Also Evident That These Traditional Methods Are Time-consuming And Include A High Possibility Of Error On The Part Of Humans. The Aim Of This Research Is To Propose An Intelligent Method Of Fruit Freshness Detection And Quality Evaluation Using Machine Learning And Image Processing Techniques. In This Context, It Is Evident That A Digital Image Of A Fruit Is Used As An Input To The System And Is Subjected To Various Preprocessing Techniques Like Resizing, Normalization, And Denoising. Read more Read less

AI BASED SAFETY MONITORING SYSTEM FOR MENTALLY EXHAUSTED ADULT WOMEN

Creative Methods Will Need To Be Developed To Keep Women Who Are Mentally Fatigued Secure, Beyond Traditional Emergency Response Systems. Cognitive Fatigue Can Impair Judgment, Slow Reaction Time And Make It Difficult For A User To Manually Activate... Creative Methods Will Need To Be Developed To Keep Women Who Are Mentally Fatigued Secure, Beyond Traditional Emergency Response Systems. Cognitive Fatigue Can Impair Judgment, Slow Reaction Time And Make It Difficult For A User To Manually Activate A Safety Alert, Making Traditional Methods Ineffective. Therefore, An AI Safety Monitoring System Has Been Developed To Operate Continually Behind The Scenes While Integrating Geospatial Intelligence With Behavior Analysis. This System Will Allow For A Continual Monitoring Of Individuals Via Their Real-time Location, Use Of Geo-fencing Technology To Create Secure Zones And Will Use Movement Pattern Analysis To Identify When Individuals Have Moved Outside Of Their Established Safe Zone And/or Exhibit Any Unusual Trends (movement Patterns) That Deviate From Their Established Behavior. Every Situation Is Categorized As Either "Safe", "Attention" Or "High Risk," Allowing For Proactive Intervention Prior To An Incident Occurring. When An Individual Finds Themselves In A High-risk Situation, The Application Will Independently Send Out Alerts And Provide The Current Location Of That Individual To The Assigning Guardians Without Relying Upon The User To Provide Input. In Addition To Providing Notifications After An Incident Occurs, The Predictive Model Of The Solution Will Enable It To Evaluate Time, Space, And Behavior In Order To Assess Imminent Threats. This Will Provide Personal Safety With A Systems-based, Adaptive Approach. Ultimately The Proposed Approach Will Include Context Awareness, Constant Surveillance, And Automatic Decision Making To Provide A Robust Support System For At-risk Individuals. The Proposed Solution Represents Acompletely New Paradigm In Personal Safety By Leveraging Artificial Intelligence, Behavioral Analytics, And Geolocation Technologies.Because The Framework Can Be Tailored To Each Person's Unique Behavioral Patterns, It Increases Trust And Dependability While Maintaining Autonomy. This Study Highlights How Intelligent, Automated Monitoring Systems Might Transform Personal Safety Paradigms By Providing Proactive, Scalable Methods For Reducing Hazards Related To Situational Vulnerability And Mental Fatigue. Read more Read less

Formulation And Evaluation Of Herbal Suncreen Cream Using Butterfly Pea Flower

Sunscreen Is A Chemical Compound That Help Protect You From UV Rays Sunburn Is Caused By Ultraviolet B Radiation But Ultraviolet A May Be More Damaging To The Skin. Sunscreen Should Ideally Block Both Wavebands. The Aim Of This Study Was To Develop... Sunscreen Is A Chemical Compound That Help Protect You From UV Rays Sunburn Is Caused By Ultraviolet B Radiation But Ultraviolet A May Be More Damaging To The Skin. Sunscreen Should Ideally Block Both Wavebands. The Aim Of This Study Was To Develop Herbal Topical Sunscreen Formulation Based On Some Fixed Oils , In Combination With Some Medical Plants. Regular Use Of Sunscreen Reduces The Development Of Actinic Keratosis , Squamous Cell Carcinoma And Melanoma . Sunscreen May Be Organic Or Inorganic Chemicals . Sunscreen Is Also Known As Sunblock Lotion. The Product That Absorb Or Reflect The Suns Ultr Aviolet Radiation And Protect The Skin. The Increasing Incidence Of Skin Cancers And Photo Damaging Effects Caused By Ultra - Violet Radiation Has Increased The Use Of Sunsreening Agents, Which Have Shown Beneficial Ef- Fects In Reducing The Symptoms . Sunsreening Agents Should Be Safe Chemically Inert , Non Irritating Non Toxic , Photo Stable An Able To Provide Complete Protection To The Skin Against Damage From Solar Radiation. Read more Read less

FIM Intelligent File Integrity Monitoring System With Ransomware Detection

File Integrity Monitoring Has Become An Essential Component Of Modern Cybersecurity Systems As Organizations Increasingly Rely On Digital Infrastructure To Store And Process Critical Data. Unauthorized Modification Of Files And Ransomware Attacks Can... File Integrity Monitoring Has Become An Essential Component Of Modern Cybersecurity Systems As Organizations Increasingly Rely On Digital Infrastructure To Store And Process Critical Data. Unauthorized Modification Of Files And Ransomware Attacks Can Lead To Severe Data Loss, Operational Disruption, And Financial Damage. Traditional File Monitoring Systems Primarily Focus On Detecting File Changes But Often Fail To Analyzebehavioral Patterns Associated With Malicious Activities. This Paper Presents An Intelligent File Integrity Monitoring System With Integrated Ransomware Detection Capabilities Designed To Improve Host-based Security Monitoring. The Proposed System Continuously Monitors File System Activities Such As File Creation, Modification, And Deletion In Real Time. A Behavioral Profiling Mechanismalong With A Hybrid Risk Scoring Model Is Applied To Analyze Abnormal Activity Patterns And Classify Potential Threats. To Detect Possible Ransomware Encryption Behavior, Entropy-based Analysis Is Performed On Modified Files. The Framework Also Incorporates Directory Traversal Detection And Multi File-type Monitoring To Identify Suspicious File Access Patterns That Commonly Occur During Ransomware Attacks. When Abnormal Activity Is Detected, Automated Response Mechanisms Including Email Alerts And File Quarantine Are Triggered To Mitigate Potential Damage. The System Also Provides A Graphical Dashboard For Real-time Monitoring And Generates Structured Security Reports For Forensic Analysis. Experimental Testing Demonstrates That The Proposed Monitoring System Effectively Detects Abnormal File Behavior And Enhances The Ability To Identify Potential Ransomware Activities In Host-based Environments. Read more Read less

Work-Life Balance And Job Satisfaction In The Gig Economy

The Rapid Growth Of The Gig Economy Has Transformed Traditional Employment Structures, Offering Flexibility While Also Raising Concerns Regarding Work-life Balance And Job Satisfaction. This Study Examines The Relationship Between Work-life Balance A... The Rapid Growth Of The Gig Economy Has Transformed Traditional Employment Structures, Offering Flexibility While Also Raising Concerns Regarding Work-life Balance And Job Satisfaction. This Study Examines The Relationship Between Work-life Balance And Job Satisfaction Among Gig Workers. A Structured Questionnaire Was Administered To 50 Respondents Engaged In Gig-based Work. Statistical Tools Such As Reliability Analysis, Descriptive Statistics, Correlation, And Regression Were Applied. The Findings Indicate A Strong Positive Relationship Between Work-life Balance And Job Satisfaction. While Flexibility Enhances Satisfaction, Irregular Income And Job Insecurity Negatively Affect Overall Well-being. The Study Highlights The Need For Policy Frameworks To Support Gig Workers’ Work-life Integration. Read more Read less

Buy Now Pay Later (BNPL) Adoption Among Youth In Kerala A Behavioural And Financial Perspective

Buy Now Pay Later (BNPL) Services Are Transforming Consumer Credit Behaviour, Particularly Among Youth. This Study Examines The Determinants Influencing BNPL Adoption Among Youth In Kerala And Evaluates Its Impact On Spending Behaviour. Using A Struc... Buy Now Pay Later (BNPL) Services Are Transforming Consumer Credit Behaviour, Particularly Among Youth. This Study Examines The Determinants Influencing BNPL Adoption Among Youth In Kerala And Evaluates Its Impact On Spending Behaviour. Using A Structured Questionnaire, Primary Data Were Collected From 250 Respondents And Analysed Using SPSS. The Findings Reveal That Perceived Usefulness, Ease Of Use, And Social Influence Positively Impact BNPL Adoption, While Financial Literacy And Risk Perception Negatively Influence Impulsive Usage. Regression Analysis Shows That BNPL Usage Significantly Predicts Changes In Spending Behaviour. The Study Provides Empirical Insights Into Fintech Adoption In A Regional Context. Read more Read less

AN ANALYTICAL STUDY ON GRIEVANCE OF THE EMPLOYEES IN CLASSIC APPAREL FASHIONS.PVT.LTD

Employee Grievances Significantly Influence Organizational Performance, Employee Satisfaction, And Workplace Harmony. This Study Analyzes The Nature, Causes, And Impact Of Employee Grievances At Classic Apparel Fashions, A Leading Textile Manufacturi... Employee Grievances Significantly Influence Organizational Performance, Employee Satisfaction, And Workplace Harmony. This Study Analyzes The Nature, Causes, And Impact Of Employee Grievances At Classic Apparel Fashions, A Leading Textile Manufacturing Firm. Using Structured Data Analysis And Employee Feedback, The Study Identifies Key Grievance Factors Such As Working Conditions, Communication Gaps, Compensation Concerns, And Work-life Balance Issues. The Findings Reveal That Ineffective Grievance Redressal Mechanisms And Lack Of Awareness Contribute To Dissatisfaction. The Study Concludes With Strategic Recommendations To Improve Grievance Handling Systems, Thereby Enhancing Employee Morale And Productivity. Read more Read less

ENVIRONMENTAL RISK INTELLIGENCE SYSTEM WITH SITUATION-BASED ALERT PRIORITIZATION

Environmental Hazards Like Extreme Temperatures, Fire, Toxic Gases, And Sudden Rainfall Threaten Safety And Operations. This Project Introduces An Environmental Risk Intelligence System For Real-time Monitoring, Analysis, And Prioritized Alerting. It... Environmental Hazards Like Extreme Temperatures, Fire, Toxic Gases, And Sudden Rainfall Threaten Safety And Operations. This Project Introduces An Environmental Risk Intelligence System For Real-time Monitoring, Analysis, And Prioritized Alerting. It Integrates Sensors For Temperature, Humidity, CO2, Rainfall, And Fire Detection, Processed By A Microcontroller Using Threshold-based Logic To Assess Risk Levels. Upon Detecting Anomalies, It Triggers Local Alerts And Sends Data To A Cloud Platform For Remote Dashboard Visualization, Storage, And Notifications. Combining IoT Sensors, Embedded Processing, And Cloud Connectivity, The System Boosts Situational Awareness And Risk Mitigation. Ideal For Smart Buildings, Industries, Agriculture, And Public Spaces, It Enhances Safety, Speeds Decision-making, And Advances Intelligent Monitoring. Read more Read less

AI-DRIVEN SECURE NETWORK AUTHENTICATION WITH SPATIAL TRUST VALIDATION

This Paper Presents An AI-Driven Secure Network Authentication System With Spatial Trust Validation To Enhance Traditional Login Security Using Machine Learning And Generative AI Techniques. During Registration, The System Stores Baseline Authenticat... This Paper Presents An AI-Driven Secure Network Authentication System With Spatial Trust Validation To Enhance Traditional Login Security Using Machine Learning And Generative AI Techniques. During Registration, The System Stores Baseline Authentication Features Including IP Address And Geolocation Along With User Credentials. During Login, Real-time IP And Location Data Are Compared With Stored Values To Detect Anomalies.An Intrusion Detection System Based On The XGBoost Algorithm Analyzes Behavioral Features Such As Login Frequency, IP Deviation, Geolocation Variance, And Failed Attempts To Classify Access Requests And Generate A Dynamic Risk Score. Suspicious Activities Trigger Instant Email Alerts And Are Logged For Analysis. Additionally, A Generative AI Module Provides Adaptive Security Recommendations. The System Establishes A Smart And Self-learning Cyber Security Framework Suitable For Modern Web Applications.The Integration Of Spatial Validation With Behavioral Analysis Significantly Reduces Unauthorized Access Risks. The Proposed Framework Supports Real-time Monitoring And Dynamic Decision-making To Strengthen Overall Network Security. Experimental Evaluation Demonstrates Improved Detection Accuracy And Reduced False Positives Compared To Conventional Authentication Systems. Read more Read less