Fuzzy logic application for house price prediction

Sarip, Abdul Ghani and Hafez, Muhammad Burhan (2015) Fuzzy logic application for house price prediction. International Journal of Property Sciences, 5 (1). (doi:10.22452/ijps.vol5no1.3).

Record type: Article

Abstract

Various methods have been used previously to estimate housing sales prices to model the underlying non-linearity relation between housing attributes and the price of property. More advanced non-linear modelling techniques such as Artificial Neural Networks (ANN), Fuzzy Inference System (FIS) and Fuzzy Logic (FL) emerged recently to model the nonlinear relation between the independent variables and the price function. A new structured model for house prices prediction based on Fuzzy Logic is proposed. A fuzzy logic based regression model has proved to be effective to address many prediction problems used in business forecasting, marketing and insurance. This paper highlights the development of a theoretical formulation for sales price prediction through the utilisation of a fuzzy regression model by applying fuzzy logic and fuzzy inference system techniques. The results show favourable outputs which indicate superior prediction function when compared with ANN and FIS as well as indicate the fuzzy functional relationship between dependent and independent variables.

Text

document 2 - Accepted Manuscript

Available under License University of Southampton Accepted Manuscript Licence.

Download (611kB)