Refurbished Car Price Prediction Using Machine Learning

In the contemporary automobile market, the purchase of a used car presents numerous challenges for buyers, chief among them being the accurate prediction of the vehicle's price. This essay uses a wide range of information, such as miles per gallon (mpg), model, actual price, year of manufacture, fuel type, gearbox mode, distance driven, tax, and engine size, to address the practical issues faced by consumers when estimating the cost of secondhand cars. Predictive models are created using a variety of machine learning techniques, such as decision tree regressor, random forest regressor, extra trees regressor, randomized search CV on random forest regressor, and regression. It is found via extensive testing that the random forest regressor model constantly yields the best accuracy. Furthermore, applying the Cat Boost method results in an impressive R-squared value of 96.21%, highlighting its efficacy in precisely forecasting used car values. By helping consumers make well-informed decisions based on accurate pricing projections, our research helps to improve the efficiency and transparency of the used car market.