Prediction of Used Car Price Based on Supervised Learning Algorithm

In this paper, we use machine learning algorithms to predict the price of used cars with less human intervention to make the results more objective. The method used is to preprocess the dataset through Python's Pycaret package and compare the performance of each algorithm through the algorithm comparison function, in this study Extra Trees Regressor, Random Forest Regressor performs relatively well. Finally, the algorithm was optimized by using the hyperparameter function. The results show that R2 = 0.9807 obtained from extreme random numbers is the best performance. The algorithm was obtained and validated with new data to derive the final algorithm model. When new used car data flows into the used car system, used car prices will be automatically generated by this algorithm, which will make the workflow of the used car market faster and more competitive for that used car market.