Boston House Price Prediction Based on Machine Leaning Methods

Abstract

A significant part of the U.S. economy is real estate and the housing industry. The level of housing prices influences the market's total price level to some extent. Data from Kaggle website is evaluated by using linear regression, random forest regressor and SVM regressor. There are 505 samples and the relationships between aim price with 13 different features. In terms of accuracy of the prediction, the method random forest is best method to predict house price. However, better model like XGboost could be chosen to improve prediction results.