Fiora.ai: An Ai-powered Personalized Skincare Routine And Product Recommendation System

• Author(s):

  Adesh Dasharath Ghodekar | Atharv Satyawan Gholap | Omkar Pandurang Kangane | Bhumika Shambhu Pardeshi | Dr. Shubhangi R. Patil

• Keywords:

  Artificial Intelligence, Skin Analysis, Deep Learn Ing, Computer Vision, MobileNetV2, Recommendation Sys Tem, Flask, Personalized Skincare

• Abstract:

  Skincare Has Become An Important Part Of Personal Grooming And Health Awareness, Yet Choosing The Right Prod Ucts And Routine Remains Difficult Because Skin Type, Envi Ronmental Exposure, And Product Ingredients Vary So Much From Person To Person. Most People Fall Back On Generic Ad Vice, Advertisements, Or A Slow And Costly Trial-and-error Pro Cess. This Paper Presents Fiora.AI, An AI-powered Web Ap Plication That Analyzes A User-uploaded Facial Image With A MobileNetV2-based Convolutional Neural Network To Classify Skin Type, Then Feeds That Prediction Into A Rule-based Rec Ommendation Engine That Generates A Personalized Morning And Night Skincare Routine Along With Specific Product Sug Gestions. The System Was Implemented Using Python, Flask, OpenCV, TensorFlow/Keras, And SQLite, With A Responsive HTML/CSS/JavaScript Front End. The Completed Prototype Was Evaluated Through Unit, Integration, System, Functional, Performance, And Security Testing, All Of Which Passed, And In Formal Classification Trials Placed Accuracy In The Low-to-mid Nineties. The Result Is A Fast, Low-cost, And Reasonably Accurate Digital Skincare Assistant That Can Be Extended Toward Concern Level Detection, Mobile Deployment, And Dermatologist Collab Oration In Future Iterations

---

Other Details

• Paper id:

  IJSARTV12I6105717

• Published in:

  Volume: 12 Issue: 6 June 2026

• Publication Date:

  2026-06-22

Download Article