Real estate valuation with the help of machine learning algorithms: a study of the real estate market of the city of Dubna

This work is devoted to the machine learning methods usage for on real estate data analysis in Dubna (Moscow region). The main purpose of the work is to compare the prognostic characteristics of various machine learning methods (linear regression, decision tree, random forest, gradient boosting). The database for our analysis consists of a sample of 800 apartment price records. The results of the analysis show that gradient boosting and random forest showed better results than other models in housing price modeling. In General, we conclude that machine learning methods can provide a useful set of tools for obtaining information about housing markets.