House Price Prediction by Machine Learning Technique—An Empirical Study

Depicting the price of a home is becoming crucial day by day, as the cost of land and houses rises year after year. House prices are significantly associated with features such as locality, region, house, age, and people; forecasting price of an individual house needs a lot of information. The House Price Index (HPI) is one of the standard tools for assessing house price variations. The objective of the study is to predict the price of houses with the use of various regression approaches of supervised machine learning. In the proposed study, housing data of 5000 homes in USA have been analyzed. Six models are used, namely, Random Forest Regression, XGBoost Regression, Linear Regression, Artificial Neural Network, Gradient Boosting and Ada Boosting. The results are compared, of which Linear Regression and Artificial Neural Network outperformed with R2_Score of 91.56 and 91.37% compared to the other four algorithms.