Early prediction of youth mental-health risk
Built an early prediction system using survey features related to sleep, stress, and social behavior. Processed datasets, trained four machine-learning models, and identified XGBoost as the strongest model across scenarios.