An Intelligent Machine Learning based System for Car Price Estimation

Price prediction is turning out to be helpful for customers, vendors, and producers of automobiles for a better price estimate. For instance, in the case of automobiles, customers, vendors, and producers apply price prediction in order to get a more accurate estimation of the price. These machine learning algorithms are used in the study and make up an exact prediction of car prices in a dataset containing the following features: make, model, year, mileage, engine size, fuel type, and transmission type. This paper is an attempt at the construction of a sturdy capable of predicting price in a reliable way. All of these purposes are achieved using different machines like linear regression models, decision trees, random forests, and gradient boosting machines. "For preprocessing the data, imputation of missing values was done in encoding categorical variables and normalization of numeric features." Besides, feature engineering can be used in generating other informative variables". Just like with other models, ensemble methods such as random forests and gradient boosting techniques are better among data points that enable generalizing without overfitting. Being able to optimize them by tuning the hyperparameters for the model. "This is how machine learning could change our minds on predicting car prices". This research concerns the problem of prediction about car prices using a dataset containing numerical and categorical features while avoiding overfitting generalization from learning from given new data in the dataset. The model got an R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> score of 0.9204, which means it explained 92.04% of the differences in car prices. It had an MSE of 312.45. Random Forest Regression did the best job making sure the results were very accurate and worked well for new data.