Housing Price Prediction Based on CNN

Abstract

Housing price has been one of the most concerned issues to the public all over the world. The excessive growth of housing price will affect not merely the quality of life, but also the business cycle dynamics. However, the factors influencing residential real estate prices are complex and the selection of effective features is vague, which leads to a lower accuracy in many of the traditional housing price prediction approaches. Accordingly, a novel prediction model based on CNN is proposed for prediction of housing price as well as the process of feature selection. Compared with other traditional methods, our work can obtain a better performance through experiments using actual data of property transaction.