Enhanced Used Car Price Prediction Using Machine Learning: A Comparative Study of Regression Models

The demand for accurate and efficient car price prediction systems is growing due to the rapid expansion of the used car market. This paper presents a comparative analysis of multiple machine learning models for predicting used car prices. The study utilizes a dataset containing various features such as the car’s year of manufacture, mileage, engine capacity, fuel type, transmission type, km driven, owner type, seller type, seats, brand and model. Models including Lasso Regression, MultiLayer Perceptron Regressor (MLPRegressor), Decision Tree Regressor, Gradient Boosting Regressor, Random Forest Regressor, and Extreme Gradient Boosting (XGBoost) were trained and evaluated. Performance metrics such as Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and R -squared ($\mathrm{R}^{2}$) were computed to assess model effectiveness. Among these, XGBoost achieved the highest $\mathbf{R}^{\mathbf{2}}$ value of 0.9118, indicating its superior predictive accuracy and robustness in capturing complex patterns in the dataset. This work highlights the potential of advanced machine learning techniques in streamlining price estimation processes, thereby benefiting stakeholders in the used car industry. The findings also emphasize the importance of feature selection and model tuning in achieving optimal results.