Real Estate Appraisal System Based On Gis And Bp Neural Network Research

With the growth of national attention on the real estate field and the continued investment, as an important part of the real estate industry, the real estate appraisal industry has also been rapid development. However, imperfect the current theory of the real estate appraisal, valuation and less efficient, and the valuation method is not accurate and so the development of real estate appraisal industry has some influence. Therefore, there are some theoretical and practical value in exploring the valuation theory and the method of valuation.On the basis of analyzing the current situation at home and abroad valuation of real estate, this paper combined with the current characteristics of the domestic property Valuation and methods to Establishment of valuation models with BP neural network model and the hedonic price valuation model, then make the model integrate with GIS to Establishment of property Valuation model based on professional geographic information system, in order to reduce the valuation of working time, promote scientific valuation and improve the quality of valuation. Firstly, this paper introduces the basic theory of property Valuation and basic valuation method, and describes the basic theory of artificial neural networks and the improved algorithm of BP neural network; secondly, by using BP neural network learning ability and superior processing power of the nonlinear relationship, it trained samples (trading case) by L-M improved BP neural network algorithm with MATLAB; thirdly, use test samples to verify the trained BP network and to the trained BP network based on construction of hedonic price valuation models; next, making overall system design, system design and system database module design for the real estate appraisal system based on professional valuation model, Finally, For VS2008 as development tools, C # as development language, ArcGIS9.3 as development platform, implemented the real estate appraisal system based on GIS and BP neural network. System application showed that the system is basically a valuation of the project spatial data and attribute data query, add, edit and output functions, to the use of BP network and GIS technologies supporting valuation study, and valuation for the property valuation methods provide a new and effective basis for decision making.