ML Regression Pipeline

A concise end-to-end machine learning workflow exploring four regression algorithms — Decision Tree, Random Forest, KNN, and Support Vector Regression (SVR). The project includes EDA, preprocessing, model training, and performance comparison.

🔍 Key Features

Full EDA-to-Modeling workflow

Clean data preprocessing & feature scaling

Comparison of 4 regression algorithms

Evaluation with MAE, MSE, and R²

Clear visualizations & insights

🤖 Algorithms Used

🌳 Decision Tree Regressor

🌲 Random Forest Regressor

📍 KNN Regressor

⚙️ Support Vector Regression (SVR)

📊 Summary

Random Forest delivered the most stable performance.

SVR performed well with proper scaling.

KNN was sensitive to feature distribution.

Decision Trees provided high interpretability.