About

Health Insurance Claim Risk Prediction using LightGBM

A machine learning solution that predicts the probability of insurance customers
filing health insurance claims. Built with LightGBM to handle highly imbalanced
data (95:5 ratio) and optimized for PR-AUC metric. Includes comprehensive EDA,
stratified validation, and interactive prediction system for risk assessment.

Tech Stack: Python, LightGBM, Scikit-learn, Pandas, Matplotlib
Metric: PR-AUC (Precision-Recall Area Under Curve)