BP神经网络在房屋批量评估中的应用

Application of BP Neural Network in the Mass Appraisal of Housing

文中基于房产税税基的批量评估,结合BP神经网络理论,构建了房地产批量评估流程及模型,运用昆明市近期已成交房屋数据对网络进行训练和测试。结果表明,可将训练成功的模型用于评估同类型大批量房屋的价格,因此,只要将不同地区房屋按地段、类型、用途进行分区,分别构建网络模型,则可对该地区的房产税进行尽快的评估。

Based on the mass appraisal of property tax base, combined with BP neural network theory, this paper constructed the mass appraisal process and model of real estate, and conducted the training and testing of the recent traded housing data in Kunming. The results show that the successful model can be used to assess the price of the same type of housing with large quantities. Therefore, so long as to conduct the regionalization of the houses in different region according to location, type and use, and construct network model, the assessment of the real estate tax in this region can be carried out as soon as possible.