基于BP神经网络模型的商品房价格预测研究

On the Prediction of Commercial Housing Price Based on BP Neural Network Model

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年~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 commercial housing prices, based on the analysis of the main factors affecting commercial housing prices, it is proposed to establish a forecasting model of commercial housing prices by using BP neural network, optimizing the structural parameters such as initial weights and thresholds of BP network by using Drosophila-Frog Leap algorithm, with the commercial housing prices of a city from 2000 to 2018 and the data of its main influencing factors selected as testing and training sample. The simulation results show that the BP neural network model optimized by Drosophila-Frog Leap algorithm can accelerate the convergence speed of the network and improve the accuracy of commercial housing price prediction, which can providesome reference value for the government sectors to carry out macrocontrol of housing prices and for the operation and management of real estate enterprises.