基于因子分析和极限学习机的房价预测研究  

Research on predicting house prices based on factor analysis and extreme learning machine

作　　者：李霞[1] 王菲 刘建平[1] 纪佳琪 Li Xia;Wang Fei;Liu Jianping;Ji Jiaqi(Hebei Normal University of Science&Technology,Qinhuangdao 066004,China;The Technology Innovation Center of Cultural Tourism Big Data of Hebei Province,Chengde 067000,China;Hebei Normal University for Nationalities,Chengde 067000,China)

机构地区：[1]河北科技师范学院,河北秦皇岛066004 [2]河北省文化旅游大数据技术创新中心,河北承德067000 [3]河北民族师范学院,河北承德067000

出　　处：《江苏科技信息》2021年第18期73-75,共3页Jiangsu Science and Technology Information

基　　金：国家自然科学基金,项目编号:62006069;河北省文化旅游大数据技术创新中心开放课题,项目编号:SG2019036-yb2001;河北科技师范学院博士科研基金和教学研究基金,项目编号:2020YB012,JYYB202034。

摘　　要：文章将影响房价因素归结为政府、消费者和房地产商三大方面,选取了影响房价波动的8个指标。为了更加准确地预测房价,首先利用因子分析对指标进行降维,得到两个公因子;然后,以该公因子作为极限学习机的输入,以对应的房价作为输出建立房价预测模型;最后,与其他3个模型预测结果进行对比分析。结果表明:结合因子分析的极限学习机比其他3个模型更能实现精准房价预测。This article divides the factors that affect house prices into three aspects:government,consumer and real estate dealer.Eight indices affecting house price fluctuation are selected.In order to predict the house prices more accurately,we first use factor analysis to reduce the dimension of the indices and get two common factors.Then,the common factor is taken as the input of the extreme learning machine,and the corresponding house prices are taken as the output to establish the house price prediction model.Finally,the prediction results are compared with those of the other three models.The results show that the extreme learning machine combined with factor analysis is more accurate than the other three models.

关 键 词：房价预测 极限学习机 因子分析 模型对比 

分 类 号：O234[理学—运筹学与控制论]