Predicting Insurance Charges Using Linear Regression

This project uses a real-world dataset to predict individual medical insurance costs using linear regression. We explore how demographic and lifestyle factors — such as age, BMI, and smoking status — influence charges.

📊 Key Features

• Data cleaning and exploratory analysis
• Correlation heatmap and feature encoding
• Linear regression model training and testing
• Model performance metrics: R² and MSE
• Visual comparison of actual vs predicted charges

🧪 Tools Used

• Python
• Pandas, NumPy
• Matplotlib, Seaborn
• Scikit-learn
• Jupyter / Google Colab

📁 Files Included

• Insurance_Predictor.ipynb: Final notebook
• insurance.csv: Dataset
• actual_vs_predicted.png: Visualization
• report.pdf: APA-style write-up
• requirements.txt: Project dependencies

📈 Results

• R² Score: 0.78
• MSE: 33.6 million
• Top Features: Smoking status, Age, BMI

🙋‍♂️ Author

Billy Pierre
LinkedIn