Machine Learning Models for Predicting Resale Value of Car

Estimating the accurate selling price is an important task in the complex automotive industry. In the Indian used car market, used car pricing requires a large degree of discretion, resulting in poor performance and uncertainty. This research paper aims to solve existing problems by providing a powerful tool to improve value estimation for stakeholders in the used car industry. Collect and pre-process data and then perform data analysis. Many machine learning algorithms subsequently emerged, including linear regression, decision trees, random forests, XG Boost, SVR, KNN, and Ensemble models. Evaluating the performance of each method shows that XG Boost is the best performing model with an R-squared score of 0.9373, MSE (Mean Squared Error) of 11.66 and MAE (Mean Absolute Error) of 1.35. Finally, the model was used to reach end users as a web application and Android application, providing practical solutions for the implementation of vehicle pricing strategies.