Fakefinder: Context-aware News Credibility Detection Using Nlp And Machine Learning

Abstract

With The Rapid Growth Of Digital Media And Social Networking Platforms, The Spread Of Misinformation And Fake News Has Become One Of The Most Pressing Challenges Of The Modern Era. Platforms Such As WhatsApp, Facebook, And Twitter Deliver Billions Of Messages Daily, A Significant Proportion Of Which Contain Fabricated, Manipulated, Or Misleading Content. Human Fact-checkers Cannot Scale To Address This Volume. This Paper Presents FakeFinder, A Context-aware News Credibility Detection System That Uses Natural Language Processing (NLP) And Machine Learning (ML) To Automatically Classify News Articles And Social Media Messages As Fake Or Real With An Associated Confidence Percentage. The System Implements A Complete NLP Preprocessing Pipeline — Text Cleaning, Tokenization, Stopword Removal, Lemmatization, And Custom Feature Engineering — Combined With TF-IDF Vectorization And Five ML Classifiers. Custom Features Including FEAT_HIGH_CAPS, FEAT_MANY_EXCLAIM, FEAT_FORWARD_MESSAGE, And FEAT_CREDIBLE_SOURCE Are Engineered To Capture The Distinct Linguistic Fingerprint Of Fake Content. The Best-performing Model Achieves Approximately 95% Accuracy. The System Is Deployed As A Flask-based Web Application With A REST API, Enabling Real-time Fake News Detection For Any Input Text. Experimental Results Confirm That FakeFinder Effectively Identifies Misinformation And Provides Reliable, Explainable Classification With High Accuracy

Keywords