Borusyak K.K, Munerman I.V.,
Chizhov S.S. Neural
Network Modeling in the Problem of Mass Appraisal of Non-Residential
Real Estate in Moscow. //
Economics of
contemporary Russia 2009. ¹3
(46) Ñ.80-86.
The paper analyses the use of neural networks for mass appraisal of
commercial property applied to the real property of The Moscow City
Property
Department. Special features of the factor space and problems of data
collection and preprocessing are described. General regression neural
networks
(GRNN) were first time used for appraisal purposes and made it possible
to
achieve the acceptable level of 20% relative error.
Keywords: neural
networks, GRNN, MLP, mass appraisal, property management.
Contents
Back
to home page