A Comparative analysis of machine learning models for real estate valuation

This article presents a comparative analysis of machine learning models used to forecast the market value of residential real estate. The study is based on data analysis, including quantitative and qualitative parameters of properties, such as area, location, transport accessibility, and level of repair. Nominal scaling was used to process categorical features, while correlation and factor analysis were employed to identify significant factors and eliminate information redundancy. The paper constructs and evaluates four forecasting models: linear regression, random forest, gradient boosting, and neural networks. It is shown that the random forest-based model demonstrates the best forecast quality.