Car Price Prediction Using Machine Learning Approach

Accurate prediction of car prices is an important problem in the automobile and resale market, influencing buyers, sellers, and dealers. With the rapid growth of online car-selling platforms, large volumes of data related to vehicle specifications and pricing are generated. This research proposes a machine learning-based approach for predicting car prices using Linear Regression and Lasso Regression models. The dataset includes various features such as car brand, year of manufacture, mileage, fuel type, transmission type, and engine specifications. Data preprocessing techniques including handling missing values, encoding categorical variables, and feature scaling were applied. Model performance was evaluated using metrics such as Mean Absolute Error (MAE), Mean Squared Error (MSE), and R² score. The results demonstrate that while Linear Regression provides a strong baseline, Lasso Regression improves generalization by reducing overfitting through feature selection. The proposed system offers a reliable and interpretable solution for estimating car prices in real-world applications.