摘要
针对房屋价格不断变化的特点,介绍了一种基于大数据方法采用深度置信网络结构建立的房屋价格预测模型。该模型通过对影响房屋价格的多种因素与房屋价格关联深度的学习进行预测,可以辅助进行房屋价格变化预测的研究,其所用的数据是过去多年公开的房屋市场中已经成交的1 461套房屋的价格数据,并与机器学习领域中经典的偏最小二乘、人工神经网络、支持向量机模型的对比实验预测结果进行验证。结果表明,该模型在房屋价格预测上能够取得较准确的结果,能够有效进行房屋价格预测。
Deep belief network was used in the prediction of house prices. A deep learning prediction model which can as- sist housing price research was established. The price data of 1 461 houses that have been traded in the Ames housing market published by the Kaggle contest platform were analyzed. Partial least squares, artificial neural network and support vector machine were used as comparative experiments to verify the accuracy of the proposed model. The experimental results show that the proposed model can achieve good results in housing price prediction.
作者
吕昊
LU Hao(Tianjin Nankai Construction Investment Company,Tianjin 300110,China)
出处
《天津科技》
2018年第10期79-81,共3页
Tianjin Science & Technology
关键词
房屋价格
预测
深度置信网
深度学习
housing price
prediction
deep belief network
deep learning
