Fuzzy logic methodology and multiple regressions for residential real-estates valuation in urban areas
Küçük Resim Yok
Tarih
2011
Dergi Başlığı
Dergi ISSN
Cilt Başlığı
Yayıncı
Erişim Hakkı
info:eu-repo/semantics/closedAccess
Özet
Valuation of real-estates is one of today's most popular study fields. The valuation ethods have changed due to life standard differences and locations of the countries. Nowadays, classical valuation methods are used for this aim in most countries. However, as a result of the rapid development in computer technologies, the artificial intelligence applications are now been used in the valuation of real-estates that are of great importance to the economies of the countries. In this study, the valuation of 120 residential real estates existing at the Konya, Turkey was made with fuzzy logic and multiple regression techniques using some of the properties of these real estates. The area, age, floor condition, physical properties and location of the real-estate were considered the input variables, and the market value of the estate was determined as the output variable. The real-estate valuation process executed with the Fuzzy Inference Systems (FIS) and Multiple Regression Analysis (MRA). Mamdani approach is preferred in FIS for this study. According to the results, the market prices were obtained as 87% accuracy with FIS and 83% accuracy with MRA. In addition, R 2 and RMSE values were 0.85 and 19 for FIS, 0.64 and 30 for MRA, respectively. © 2011 Academic Journals.
Açıklama
Anahtar Kelimeler
Fuzzy logic, Mamdani, Multiple regressions, Residential real-estate valuation
Kaynak
Scientific Research and Essays
WoS Q Değeri
Scopus Q Değeri
N/A
Cilt
6
Sayı
12