Automatic valuation of Jordanian estates using a genetically-optimised artificial neural network approach

摘要: Estate appraisal can be defined as valuing a land or property of given date using common data utilising standardised methods and statistical testing. has several applications such asset valuation for lenders, tax estimation, insurance estate planning, grant new mortgages to home buyers, purchase mortgage packages, which contain thousands mortgages, investments. It is also used guide potential buyers sellers with making purchasing decisions. The drawbacks on-site manual include: it time-consuming, costy, based on subjective judgments sometimes validation negotiated price rather than estimating the true market value property. In this paper, authors build an Artificial Neural Network model purpose automatic Jordanian estates avoid appraisal. proposed built in two stages: first stage, Genetic Algorithm optimiser determine best topology Network. second optimised trained values weights. evaluating price, factors are taken into consideration area size, location, lot area, establishment year, building square meter, type (commercial agricultural) others. Records from Department Lands & Survey training testing its performance find that represents underlying relation between land/property characterising features. Statistical tests were performed validate effectiveness method.