Machine Learning

Intelligent Health Insurance Cross-Sell Prediction System

The Intelligent Health Insurance Cross-Sell Prediction System is a comprehensive machine learning solution that predicts whether a health insurance customer will purchase vehicle insurance. This project is ideal for B.Tech, MCA, CSE, and engineering students as their final year project. It implements Ridge and Lasso Regression with SMOTE for class imbalance, achieving 86.4% accuracy and 0.902 ROC-AUC.

The system utilizes the Health Insurance Cross-Sell Prediction dataset from Kaggle containing 381,109 customer records with 11 features including age, annual premium, vehicle age, and previous insurance status. It provides a user-friendly web interface for data upload, exploratory data analysis, model training, and real-time cross-sell prediction, enabling insurance companies to optimize their marketing campaigns and increase revenue.

• Multi-Model Cross-Sell Prediction
• Ridge Regression (86.4% Accuracy)
• Lasso Regression (86.2% Accuracy)
• Feature Engineering (Age_Group, Premium_Group)
• SMOTE for Class Imbalance Handling
• Interactive Visualizations
• Real-time Purchase Predictions
• Model Performance Comparison
• Feature Importance Analysis
• Flask Web Application