摘要
鉴于目前商品房价格预测方法存在的问题,在分析影响商品房价格主要因素的基础上,提出采用BP神经网络建立商品房价格预测模型,利用果蝇-蛙跳算法优化BP网络初始权值和阈值等结构参数,选取某城市2000~2018年的商品房价格及其主要影响因素数据作为训练样本和测试样本.通过仿真分析表明:BP神经网络模型经过果蝇-蛙跳算法优化后能加快网络的收敛速度,提高商品房价格预测的精准度,对于政府部门进行房价宏观调控以及房产企业的运营管理都具有一定的参考价值.
In view of the problems existing in the current forecasting methods of commodity house prices, based on the analysis of the main factors affecting commodity house prices, BP neural network is used to establish the prediction model of housing price, and drosophila-leapfrog algorithm is used to optimize the initial weights and thresholds of BP network structure parameters, selection of a certain city from 2000 to 2018 of commodity house price and its main influencing factors of data as the training and test sample. The simulation analysis shows that the BP neural network model can accelerate the convergence speed of the network and improve the accuracy of the prediction of the price of commodity house after optimized by drosophila-leapfrog algorithm, which has certain reference value for the macro-control of housing price by government departments and the operation and management of real estate enterprises.
作者
乔维德
QIAO Wei-de(Department of Scientific Research&Quality Control,Wuxi Open University,Wuxi,Jiangsu 214011,China)
出处
《石家庄学院学报》
2019年第6期127-133,共7页
Journal of Shijiazhuang University
基金
常州市“831工程”后续研究项目(CZ8312012017)
无锡市社会事业领军人才资助项目(WX530/2018026YB)
关键词
BP神经网络
果蝇-蛙跳算法
商品房价格预测
BP neural network
drosophila-leapfrog algorithm
prediction of commodity house price
