Research on housing prices prediction-take Boston as an example

The problem of predicting Boston housing prices belongs to the field of artificial intelligence, specifically in the domain of regression problems, which is a crucial area of study in machine learning. This article aims to get an accurate prediction model of Boston housing prices by comparing four regression models based on the datasets derived from StabLib library. Multiple Linear Regression model, Random Forest Regression model, Extreme Gradient Boosting Regression model and Support Vector Machine Regression model are taken into consideration. Five evaluation index R-squared, adjusted R-squared, mean absolute error, mean squared error and root mean squared error are compared in terms of generalization ability of model. Eventually, Extreme Gradient Boosting Regression model was found to be the most effective model when predicting housing prices in Boston. The model has certain positive applications in real life, which can help government formulate real estate policies and people make wiser house purchasing strategies.