Artif Intell Rev
DOI 10.1007/s10462-016-9531-5
Enhancement of parcel valuation with adaptive artificial
neural network modeling
¸ Sükran Yalpır
1
© Springer Science+Business Media Dordrecht 2016
Abstract This study targets a research on the application of artificial neural network (ANN)
and multiple regression analysis (MRA) approaches in Geomatics Engineering science to
land valuation process. The prediction capability was investigated and evaluated using three
ANN models constructed with different activation functions (sigmoid, tangent hyperbolic
and adaptive activation function) and MRA was used as a reference approach. These four
methodologies were applied to land valuation in order to model the unit market value with
various inputs based on essential criteria. All approaches were investigated with their esti-
mation level in training and testing data. It was observed that adaptive ANN performed
noticeably higher predicting the values with the highest accuracy and giving the smallest
RMSE value in validation process, although other methodologies approximated to the raw
data at a promising level for further valuation-based applications.
Keywords Real estate valuation · Artificial neural network (ANN) · Adaptive ANN
1 Introduction
Real estate valuation in urban areas is a complicated and laborious task attracting growing
interest of academic research using mathematical models recently. Parcels are unresided
land portions that need to be modeled apart from residential real estates. The process of
valuation aims to estimate the worth of landed properties based on experience and judgment by
identifying and assessing their characteristics. The “value” is a term generally characterized
by some description such as market value or benefit value. The real estate market value is
defined as the estimated amount of money for which a property should exchange on the date of
valuation between a willing buyer and a willing seller in arm’s-length transaction after proper
B ¸ Sükran Yalpır
sarici@selcuk.edu.tr
1
Geomatics Engineering Department, Selcuk University, 42031 Konya, Turkey
123