stress-prediction

Explainable, counterfactual & GenAI pipeline that turns sleep & lifestyle data into actionable stress-reduction advice — Predict (CatBoost/RF/TabNet) → Explain (SHAP) → Prescribe (DiCE) → Naturalize (GPT-4o-mini).

Machine learning project for stress level and burnout prediction using XGBoost with a Flask-based frontend.

Client-side React + Vite web app to record and process voice audio and send features to an API for automated stress prediction (speech-based stress detection).

End-to-end ML pipeline to predict student mental health scores based on social media usage, sleep, stress level, and study habits. Built with Scikit-learn, XGBoost, and Streamlit.

Machine Learning framework to predict student academic stress using lifestyle and academic indicators.