Property evaluation based on ambiguous logic through building inspection in São Paulo city, Brazil

Abstract

The civil engineering branch is strongly related to the development of the countries and there is still a lot of information available in the buildings constructed. However, these data are dispersed without proper treatment. On the other hand, if these real estate data are reorganized to discover behavior parameters, these properties’ values can be predicted and still work as data and causal relationships between explanatory variables. The purpose of the research is to use construction inspection strategies associated with artificial intelligence to predict the market value of a residential apartment. In this academic experiment, only 6 samples of residential apartments are used. Those samples are located in the Lithuania Republic square at Vila Zelina neighborhood, in São Paulo, Brazil, a source in February 2021. The method uses the results of the inspection of civil engineering and converts them into linguistic terms. The result considers the imprecision, uncertainty, and subjectivity of human expression combined with artificial intelligence and civil engineering. To test the feasibility of the process, a comparison is made between the market values of the samples and the values predicted by the Fuzzy logic. Thus, the good results derived a Main Percentage Absolute Error (MAPE) of 5%, the mean absolute error (MAE), root-mean-square error (RMSE), and determination coefficient (R2) of 0.99.