Emnet-tes/AlphaCare-Insurance-Risk-Analytics

AlphaCare-Insurance-Risk-Analytics uses South African car insurance data to identify risk segments, optimize premiums, and support data-driven marketing. # 🧠 AlphaCare-Insurance-Risk-Analytics ## 🎯 Project Objective To uncover risk segments and build predictive models using South African car insurance data, enabling AlphaCare Insurance Solutions to: - Optimize insurance premiums - Identify low-risk customers for potential premium reduction - Improve marketing strategy using data-driven insights --- ## ✅ Tasks Completed ### Task 1: Git & Exploratory Data Analysis - Set up GitHub repository with version control and branching - Performed detailed EDA to understand: - Risk patterns (loss ratios) by region, vehicle type, and customer demographics - Outliers and distributions in financial/vehicle features - Temporal trends in claims and premiums - Top/bottom vehicle models by claim amount - Generated 3+ insightful visualizations ### Task 2: Data Version Control (DVC) - Installed and configured DVC for reproducible pipelines - Tracked raw dataset using DVC - Integrated DVC with GitHub using `.dvc` and remote storage ### Task 3: A/B Hypothesis Testing - Formulated and tested 4 business-critical null hypotheses: - Risk differences across provinces, zipcodes, gender - Margin differences across zipcodes - Applied statistical tests (t-test, chi-square) - Interpreted p-values to drive actionable business insights ### Task 4: Predictive Modeling - Built ML models for: - Claim severity (`TotalClaims`) prediction - Premium optimization - Optional: Claim probability classification - Models used: Linear Regression, Random Forest, XGBoost - Evaluated models using RMSE, R², and classification metrics - Applied SHAP for model interpretation and feature impact analysis --- ## 🧪 Environment Setup ```bash # Clone the repository git clone github.com # Navigate to project folder cd AlphaCare-Insurance-Risk-Analytics # Create and activate virtual environment python -m venv venv source venv/bin/activate # On Windows: venv\Scripts\activate # Install dependencies pip install …