README.md

Drug-Food Interaction Predictor Machine learning system for predicting drug-food interactions with explainable AI. Identifies potentially harmful medication-food combinations and provides risk assessments with mechanistic explanations.

Features 8 ML Models: LightGBM, XGBoost, CatBoost, Random Forest, Extra Trees, Gradient Boosting, MLP, Voting Ensemble Risk Classification: Automatic HIGH/MODERATE/LOW categorization Explainable AI: SHAP and LIME analysis for model interpretability Web Interface: Interactive dashboard with real-time search REST API: JSON endpoints for programmatic access

Usage Web Interface Select medication from dropdown Select food item Click "Analyze Interaction" View risk level, mechanism, and recommendations

Quick Start

Clone repository

git clone https://github.com/yourusername/drug-food-interaction-predictor.git cd drug-food-interaction-predictor

Install dependencies

pip install flask pandas numpy scikit-learn matplotlib seaborn pip install lightgbm xgboost catboost shap lime joblib

Train model (optional - pre-trained model included)

python main.py

Run web application

python app.py

Disclaimer ⚠️ This is a research prototype for educational purposes only. Not validated for clinical use. Always consult healthcare professionals before making medication decisions.