Fuel Efficiency (MPG) Prediction Using Machine Learning

Fuel efficiency plays a crucial role in automotive design, environmental sustainability, and performance analysis. This project presents a Machine Learning approach for predicting Miles Per Gallon (MPG) using vehicle features from the well-known Auto MPG dataset available at the UCI Machine Learning Repository.The dataset undergoes pre-processing steps including handling missing values, converting data types, and selecting key numerical attributes. Two predictive models-Linear Regression and Random Forest Regressor-are implemented and evaluated using standard regression metrics such as Mean Absolute Error (MAE), Mean Squared Error (MSE), and R² score. The Random Forest model performs significantly better, indicating its strength in capturing nonlinear patterns in vehicle characteristics.The study highlights the potential of Machine Learning to support automobile efficiency analysis and fuel consumption forecasting. Future enhancements may include model tuning, advanced algorithms, real-time prediction systems, and deployment through a web interface.