House selling price assessment using two different adaptive neuro-fuzzy techniques

• I demonstrated the utility of the ANFIS in the estimation of house selling price.
• I used two different ANFIS models.
• The models are ANFIS with grid partition and ANFIS with sub clustering.
• The ANFIS models proved to be sufficiently accurate.
• The study showed that performance of the ANFIS-GP model was better than ANFIS-SC.

It is of high importance to estimate the house selling prices because sellers and buyers need this information in an accurate way at time of sale. As a result of the fact that the estimation of the house prices is extorsive considerable research has been published in the literature on estimating house prices. This study covers the application of two different adaptive neuro-fuzzy (ANFIS) approaches for the estimation of house selling price.In this study, two different ANFIS models, namely ANFIS with grid partition (ANFIS-GP) and ANFIS with sub clustering (ANFIS-SC), were used in forecasting house prices, and the results were evaluated. Comparison of results from the two techniques indicated that the ANFIS-GP models performed better than the ANFIS-SC models. The study suggests that the ANFIS-GP technique can be successfully used in the estimation of house prices in the construction sector.