Linear Regression with Scikit-Learn: From Basics to Healthcare Cost Prediction

This work demonstrates how to apply linear regression using Python and scikit-learn, starting from a simple synthetic example and progressing to a real-world healthcare dataset.

What it Covers

Part 1: Linear Regression Basics

• Model training and prediction
• Evaluation using Mean Squared Error and R² Score
• Plotting regression line

Part 2: Medical Cost Prediction (Mini Project)

• EDA
• Feature selection and preprocessing
• One-hot encoding of categorical variables
• Splitting into train/test sets
• Model training and evaluation
• Actual vs Predicted cost visualization

Techniques & Tools Used

• Python
• scikit-learn (LinearRegression, train_test_split, metrics, OneHotEncoder)
• pandas for data handling
• matplotlib and seaborn for visualization

Next Steps

• Extend to Polynomial Regression
• Try regularization (Ridge, Lasso)
• Explore healthcare datasets with more features or classification tasks