商务大数据环境下特定类型用户智能识别方法  

Intelligent Identification of Specific Type of Users within Business in the Era of Big Data

作　　者：王斐然 姜磊[1] 文一凭[1] 周倩 Wang Feiran, Jiang Lei, Wen Yiping, Zhou Qian(Key Laboratory of Knowledge Processing and Networked Manufacturing, Hunan University of Science and Technology, Xiangtan 411201, China)

机构地区：[1]湖南科技大学知识处理与网络化制造实验室

出　　处：《湖南科技大学学报（自然科学版）》2018年第3期80-85,共6页Journal of Hunan University of Science And Technology：Natural Science Edition

基　　金：国家自然科学基金资助项目(6177219);湖南省教育厅重点资助项目(15A064);湖南省教育厅创新平台开放基金资助项目(17K033);湖南省自然科学基金资助项目(2016JJ2056;2017JJ4036;2018JJ2139)

摘　　要：随着大数据时代的来临,商家拥有了越来越多的数据,从而能够准确识别出用户类型.准确的用户类型识别可以帮助商家进行精准营销,提高销售利润.由于商务数据存在数据不全等问题,若直接应用现有的分类方法识别用户类型,识别效果并不太好.借鉴关联规则算法的思想,设计一种商务大数据环境下可智能识别特定类型用户的方法.该方法首先选取种子商品,然后利用种子商品和其他商品的强关联规则,分层次赋予商品权重,再将权重映射到用户身上从而得到一个特征值,最后依据用户特征值来识别用户类型.该方法只利用购物记录数据,就能有很好的表现.通过真实数据集上的实验,以识别超市中婴幼儿家庭用户为例,证明了该方法的效果.With the advent of the era of big data,merchants have more and more data,which accurately identify the user type. Accurate user type identification help merchants carry out accurate marketing and increase sales profit. Due to the problem of incomplete data in business data,it was not very good to identify user types directly by existing classification methods. Based on the idea of association rules algorithm,a method for intelligent identification of specific types of users in a business big data environment was designed. First a seed of goods was elected in this method,and strong association rules of seed and other commodities were used to give weight to the commodity. Then the weight was mapped to users to get an eigenvalue. Finally,the user type was identified by the user ＇s eigenvalues. This method performed well when only using shopping to record data. In order to identify the infant family in the supermarket as an example,the results of this method were verified through experiments on real data sets.

关 键 词：关联规则 用户类型 智能识别 分类 商务 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]