Estudo dos condicionantes espaciais para avaliação imobiliária utilizando técnicas de inteligência artificial – São Paulo/SP

• Ano de defesa: 2021
• Autor(a) principal: Honda, Felipe Pereira
• Orientador(a): Melanda, Edson Augusto
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de São Carlos; Câmpus São Carlos
• Programa de Pós-Graduação: Programa de Pós-Graduação em Engenharia Urbana - PPGEU
• Palavras-chave em Português: Mineração de dados; Inteligência artificial; Valorização imobiliária
• Palavras-chave em Inglês: Data mining; Artificial intelligence; Real estate valuation
• Área do conhecimento CNPq: ENGENHARIAS
• Link de acesso: https://repositorio.ufscar.br/handle/20.500.14289/14587

The increase in the supply of real estate in recent years has influenced the increase in credit, and, however, has raised sales prices. However, the measurement of real estate values is still a very complex process, as it is influenced by numerous factors, among them physical, geolocation and political-economic. Thus, the objective of this research is to present which variables are relevant in the formation of the value of urban properties using artificial intelligence technique. It was decided to use the decision tree technique for mining and statistical analysis, created from the Randon Forest algorithm. Through mining it was possible to notice that there is a strong correlation between the categories and subcategories analyzed, and, however, the influence they have on the formation of the value of urban properties in the 32 administrative regions of the municipality. Among the factors that had the greatest influence on the value of the properties were the proximity to Shopping Centers, police stations, areas where geological hazards occur, slums, consulates, fire departments and train stations. It was also found that the different environmental aspects, positive and negative, influence the real estate valuation and that the adopted methodology is efficient in the evaluation of large databases, as well as an important tool for decision making and for urban planning.