Project Overview
This project demonstrates a predictive risk model for assessing the likelihood of developing chronic conditions based on demographic and lifestyle factors. The model combines logistic regression with feature engineering and is visualized through an interactive risk calculator.
Key Features
- Data preprocessing & cleaning pipeline
- Interpretability with SHAP values
- Real‑time risk score calculator
- Responsive visualizations of model performance
Interactive Risk Calculator
Model Performance
Our model achieves an AUC‑ROC of 0.86 on a hold‑out test set, with calibration plots indicating reliable probability estimates.
