Comparison of models for predicting Real Estate Prices on Real Data

The aim of this bachelor thesis is the use of web scraping techniques to obtain real data, application of selected regression models for estimation of property price and subsequent comparison of these models. The thesis is divided into theoretical and practical parts. The theoretical part introduces the field of web scraping, including ethical and legal aspects of web scraping. Furthermore, the domains of data mining, machi ne learning, neural networks and regression analysis principles are described. For each regression model, the theoretical basis, advantages, disadvantages, applicability and limitations in the real estate market are described. The practical part includes the application of web scraping, subsequent creation and preprocessing of dataset, use of individual prediction models, their evaluation and visualization. After the practical part of the paper, the individual results are discussed. Differences in the performance of each model, inter-comparisons and discussion of the suitability for practical use are discussed. Finally, the whole work is summarized, key contributions are highlighted, and possible future extensions are outlined.