Car Price Prediction Using an Artificial Neural Network

Abstract

When buying and selling cars, it can be a challenge to assign the correct price. Artificial neural networks, a branch of artificial intelligence, are frequently used for such calculations. In this study we designed two different artificial neural networks for car price forecasting and tested them using data from a car sales website. For data, a software was developed using the C# programming language and the MSSQL Server database management system, also HTMLAgilityPack library was used to read the data on the website. A procedure was written with T-SQL language to digitize the data. As a result of the study, using data from approximately 1000 cars, a success rate of 91.38% was obtained. More data is needed for better results.