Investigation of the eTS Evolving Fuzzy Systems Applied to Real Estate Appraisal
Tadeusz Lasota, Zbigniew Telec, Bogdan Trawinski and Krzysztof Trawinski
The evolving Takagi-Sugeno algorithm (eTS) was employed to built models assisting with property valuation on the basis of real-world data drawn from a cadastral system, registry of sales transactions, and cadastral map. The eTS performance was compared with evolutionary fuzzy systems implemented in KEEL. Statistical analysis of the results was performed using nonparametric tests followed by post hoc procedures designed for multiple comparisons. The results confirmed the advantages of the eTS algorithm.
Keywords: Evolving fuzzy systems, eTS learning algorithm, property valuation, feature selection, GIS, KEEL.