非线性属性置信规则测度的茶叶消费者偏好挖掘模型  

Tea Consumer Preference Mining Model with Nonlinear Attribute Confidence Rule Measure

作　　者：袁薇[1] 唐军芳[2] 

机构地区：[1]浙江树人大学现代服务业学院,杭州310000 [2]浙江树人大学信息科技学院,杭州310000

出　　处：《科技通报》2017年第6期167-169,261,共4页Bulletin of Science and Technology

摘　　要：为了精确的衡量消费者的购物偏好,为企业提供决策信息,本文在大数据关联规则挖掘技术的基础上,建立了一种基于非线性属性置信规则测度的茶叶消费者偏好挖掘模型。首先根据非线性属性的因果关系构建置信规则库,然后以此建立消费者偏好挖掘模型,采用加权乘法聚集函数计算模型的逻辑关系,并针对模型的误差影响因子对其偏好等级的信度分配参数进行优化。模型实例仿真结果表明,与茶叶实际销售数据进行对比,本文提出的模型得到的最终结果与实际结果基本一致,证明了该模型的有效性和实用性。In order to accurately explore the shopping preference of consumers and provide decisionmaking information to enterprises, a tea consumer preference mining model based on confidence rules with nonlinear attributes is built in this thesis by virtue of large data association rule mining technology.Firstly, a confidence rule base is constructed according to the causal relation of nonlinear attributes.Secondly, a consumer preference mining model is established, and the logical relation of the model is calculated by using the weighted multiplicative aggregation function. Thirdly, the confidence allocation parameters of the preference rating are optimized aiming at error influence factor of the model. The model instance simulation results show that compared with the actual sales data of tea, the final results obtained by this model are basically the same as the actual results, and the validity and practicability of the model are proved.

关 键 词：关联规则 茶叶销售偏好 置信规则 非线性属性 加权惩罚聚集函数 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]