基于神经网络的宗地地价评估探析  

THE EVALUATION OF LAND-VALUE BASED ON ARTIFICIAL NEURAL NETWORKS

作　　者：王胜武[1] 郑新奇[1] 

机构地区：[1]山东师范大学人口.资源与环境学院,济南250014

出　　处：《山东师范大学学报（自然科学版）》2006年第4期74-78,共5页Journal of Shandong Normal University(Natural Science)

基　　金：山东省自然科学基金项目(Y2204E04)

摘　　要：当前,人工神经网络的研究方兴未艾,已成为人工智能领域的研究热点,其应用领域也越来越广.根据宗地地价评估的特点,对人工神经网络在宗地地价评估中的应用进行了探索.首先分析了宗地地价的特点和影响因素,并选取了济南市住宅用地交易样点,然后在DPS软件中用神经网络对所选样点进行宗地地价评估.通过预测结果与实际结果的对比得出,基于神经网络的宗地地价评估简便可行,并比传统方法更客观.Now the research of artificial neural networks is in the ascendant, which has become the research hotspot in the artificial intelligence, whose application has become more and more wide. The application of artificial neural networks in land - value evaluation is studied according to the characteristic of the land - value evaluation. First, it analyses the characteristic of land - value evaluation and chooses some dealing samples in Jinan landmarket, then it applies artificial neural networks in land - value evaluation. The result shows that the artificial neural networks can apply in land-value according to the contrast of the forecast value and practical value. Moreover, this method is more impersonal, simple and convenient than tranditional one.

关 键 词：神经网络 宗地 地价评估 

分 类 号：F293.2[经济管理—国民经济]