摘要
商品住宅的合理定价问题是开发商开盘前面临的重大课题。文章运用BP神经网络非线性的学习能力,通过商品住宅定价的流程设计,运用商品住宅特征价格理论,分析商品住宅价格的影响因素,并采用大量样本数据进行模拟训练与验证,模型结果显示交易样本输出值与实际值相对误差在10%以内,达到精度要求。文章探索将BP神经网络引入到商品住宅定价中,避免了传统定价方法主观性较强的弱点,同时还为批量定价或估价提供了一种新思路。
The pricing of commercial housing is a major issue for the developers before the sale.On the basis of hedonic price,through the process design of the commercial housing pricing,and with the use of BP neural network nonlinear ability,the paper analyzes the influencing factors,and establishes an evaluation system of the pricing of commercial housing.A large number of sample data have been used to have the model practiced and validated.The results show that the relative error of the output value and actual value is less than 10%,which means it reaches the accuracy requirements;thereby we establish the pricing BP neural network model based on commercial residential buildings in Hangzhou.This study avoids the subjective weaknesses of traditional pricing methods.It fits more with the nonlinear relationship between the characteristic variables and commercial housing,and provides a new way of thinking to the volume pricing or valuation.
出处
《浙江工业大学学报(社会科学版)》
2012年第3期337-341,共5页
Journal of Zhejiang University of Technology:Social Sciences
关键词
BP神经网络
商品住宅
定价
BP natural network
commercial housing
pricing