童鞋外观类别与顾客偏好度关联分析  

Analysis of association between children's shoes appearance categories and customer preference degree

作　　者：胡文超[1] 周余庆[1] 李峰平[1] 陈童[2] 叶婷[1] 

机构地区：[1]温州大学,浙江温州325035 [2]温州职业技术学院,浙江温州325035

出　　处：《中国皮革》2016年第5期52-56,共5页China Leather

基　　金：浙江省公益性技术应用研究计划项目(2012C21086);国家自然科学基金资助项目(71101112);浙江省留学人员科技活动小额资助项目(R20150102)

摘　　要：针对童鞋外观类别与顾客偏好度的关联分析,提出了一种基于结构熵聚类与多类Logistic相结合的关联模型。首先,选择若干目前流行的童鞋产品,进行童鞋外观要素提取,采用结构熵聚类方法,对诸多外观要素进行聚类分析,以获得主流的外观类别群体。然后,采用多类Logistic方法,建立顾客个体特征与童鞋外观类别的关联模型。该模型可以在顾客网络购鞋时,自动找出与顾客特征最匹配的童鞋外观类别,实现主动推荐功能。最后,实例分析说明了本方法的有效性。According to the association problem of children＇s shoes appearance categories and customer preference, an associa- tion model based on structure entropy clustering and multi - class Logistic methods was provided. Some popular children＇s shoes were chosen to extract essential factors of product appearance, and these factors were clustered by structure entropy clustering method to obtain mainstream appearance categories. The network survey data was used, which optimized multi - class Logistic as- sociation probability for the characteristics of individual customers and the appearance categories of children＇s shoes, and an associ- ated model connecting customer characteristics and children shoes＇ appearance categories was obtained. The proposed model could find the appearance categories that can most match customer characteristics in the network purchase circumstances, and provide ef- fective method guidance for personalized recommendation. Finally, an example shows the effectiveness of the proposed model.

关 键 词：童鞋外观 顾客偏好度 主动推荐 结构熵聚类 多类Logistic 

分 类 号：TS943.723[轻工技术与工程—服装设计与工程]