节点文献

基于BP神经网络模型的住宅用地地价评估研究——以南昌市主城区为例

Urban Residential Land Price Evaluation Based on BP Neural Network and GIS——Taking Nanchang City as an Example

  • 推荐 CAJ下载
  • PDF下载
  • 不支持迅雷等下载工具,请取消加速工具后下载。

【作者】 陈志鹏郭熙赵丽红郭细根

【Author】 CHEN Zhipeng;GUO Xi;ZHAO Lihong;GUO Xigen;College of Land Resources and Environment,Jiangxi Agricultural University;College of Economics and Finance,Chongqing University of Technology;

【通讯作者】 赵丽红;

【机构】 江西农业大学国土资源与环境学院重庆理工大学经济金融学院

【摘要】 减少土地估价过程中的人为主观性对于提高土地估价的科学性和客观性具有重大意义。以南昌市主城区为研究区,选取100多个住宅用地交易案例为样本,运用BP神经网络和GIS空间分析方法,分析住宅地价及其影响因子之间的复杂关系,构建BP网络建立地价与影响因子之间非线性映射关系并进行地价预测。研究结果表明:商业因素、公共设施因素、交通因素和环境因素对住宅地价有显著的影响;用BP神经网络对地价的预测误差均控制合理范围以内,BP神经网络强大的非线性映射能力和自主网络学习,理论上可以实现对住宅地价的评估,为政府和经济主体参与经济活动提供科学的依据。

【Abstract】 This study takes residential land in urban areas of Nanchang as the research object.According to the 2016 Nanchang benchmark land price evaluation report,103 trading cases were selected as sample points.This paper briefly describes the calculation principle of BP neural network,and uses BP neural network and GIS to analyze the complex relationship between impact factor and residential land price.The study found that commercial factors,infrastructure factors,traffic factors and environmental factors have a significant impact on residential land prices.GIS’s powerful spatial analysis capabilities can quantify the influencing factors.At the same time,BP neural networks have strong nonlinear mapping capabilities.Training and simulation of impact factors can theoretically evaluate the residential land price in Nanchang.The research results show that the BP neural network is used to predict the land price,and the error between the prediction result and the actual result is controlled within 1%,which indicates that the BP neural network can be applied to the urban residential land price evaluation in Nanchang city,for the government and economic subjects.Participate in economic activities to provide a scientific basis.

【基金】 江西省赣鄱英才“555”领军人才项目(201295);江西省高校人文社科项目(GL1323)
  • 【分类号】TP183;F299.23
  • 【被引频次】4
  • 【下载频次】281
节点文献中: 

本文链接的文献网络图示:

本文的引文网络