<![CDATA[Along with the development of the era of production of motorized vehicles, especially cars today, has experienced a very significant increase, companies are competing with each other in issuing the latest series. This has an impact on the abundance of used or used cars on the market, retail companies are demanded to be selective in determining the selling price and purchase price of the cars to be purchased and those to be resold, by utilizing technology, especially Data Mining, which is expected to assist the car selection process. which will be purchased quickly and accurately. Used car price predictions are influenced by several factors, including the type of car, fuel, distance traveled, year of production, and transmission. With these problems, the author tries to compare the K-Nearest Neighbors and Random Forest algorithms as the basis for making machine learning models that can predict the price of used cars according to existing specifications. The results of this study indicate that the Random Forest Algorithm has a smaller error value and better accuracy than K-Nearest Neighbors, Random Forest has an accuracy of 96.38% and K-Nearest Neighbors 59.17%, the error value obtained is calculated using the MSE (Mean Squared Error) evaluation metric by calculating the average difference between the actual value and the predicted value.]]>
Copyrights © 2023
