改进的ANFIS在房产评估中的应用  

Application of improved ANFIS in real estate property assessmen

作　　者：史东辉[1] 

机构地区：[1]安徽建筑大学电子与信息工程学院,合肥230088

出　　处：《计算机工程与应用》2014年第12期247-251,256,共6页Computer Engineering and Applications

摘　　要：房产评估是个较为复杂的非线性过程,目前的方法存在房产近邻难以定义等问题。为解决这一问题,提出基于k近邻的自适应神经模糊推理方法,并应用于房产评估。该方法通过定义不同意义的全变量、部分变量、空间、时空的k近邻,计算k近邻均价,将k近邻均价加入模型。实验结果表明,使用基于空间k近邻和时空k近邻改进自适应神经模糊推理方法对房产价格进行预测,准确性显著提高。The assessment of real estates is a complicated process in which there are some problems in the definition of neighbor. An adaptive neuro-fuzzy inference system added the k-nearest neighbors approach is used in the assessment of residential properties. By defining different k-nearest neighbors including all variables, partial variables, location, location and time, the average neighbor price of every real estate can be computed, then it is added into the input variables of the model. The simulation results show that the model using k-nearest neighbors with a location or with a location and time es-timates the prices of properties more accurately.

关 键 词：自适应神经模糊推理系统 K近邻 房产评估 地理位置 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]