聚类分析在互动电视产品销售片区划分中的应用  

The Application of Cluster Analysis in Interactive TV Selling Area

作　　者：黎永安[1] 

机构地区：[1]东方有线网络有限公司,上海200065

出　　处：《数学的实践与认识》2015年第3期15-21,共7页Mathematics in Practice and Theory

基　　金：东方有线网络有限公司数据库营销项目

摘　　要：聚类分析是数据挖掘中一种常用的技术,是根据事物本身特性来研究个体分类,按照物以类聚的原则来研究的事物分类.在商业上,聚类分析可以多角度用于市场分析,比如通过分析发现不同的客户群,并且通过购买模式刻画不同的客户群的特征,为市场营销和策略的制定提供科学合理的参考.从不同网络区域覆盖数字电视用户的一些特征(包括市场占有特征、产品消费特征等)出发,应用两步聚类方法将网络区域进行划分,得到不同的用户群体,然后针对不同的群体特征加以分析,为互动电视产品的市场营销战略和策略的制定提供科学合理的参考.Cluster analysis is a kind of common technique on data mining, which is used for making classification according to individuals＇ characteristics, in the principle of things of one kind coming together. In market analysis, cluster analysis can be used from various angles. For example, we are able to, not only afford a scientific and reasonable support to develop marketing strategies, based on customers＇ segmentation, but also group their purchasing characters. In accordance with customers characters covered by different networks （Including market shares, consumption characteristics, etc.）, this article will firstly represent the method of two-step clustering customers in marketing area, and secondly describe clusters characteristics so as to provide the support on marketing strategies about our main product-Interactive TV.

关 键 词：数据挖掘 聚类分析 互动电视产品 市场分析 客户细分 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论] G220.7-F[自动化与计算机技术—计算机科学与技术]