Advanced machine learning models analyze linguistic patterns, sentiment, and structural features to identify misinformation with high precision in real-time.
Powered by advanced natural language processing and ensemble machine learning
Instant classification of news articles using TF-IDF vectorization combined with 17 statistical linguistic features.
Soft voting ensemble of Logistic Regression, Random Forest, and SVM for robust prediction confidence.
Human-readable reasoning behind every prediction including sentiment analysis and sensational word detection.
Upload CSV files for bulk analysis with automatic aggregation of fake vs real distribution and confidence metrics.
Protected dashboard with user authentication, session management, and secure password hashing (PBKDF2-SHA256).
Optimized preprocessing pipeline with NLTK tokenization, stemming, and stopword removal for sub-second predictions.