分类算法及其在电信客户保持的应用  

Classification Algorithm and Its Application in Telecom Customer Retention

作　　者：左国才[1] 周荣华[1] 符开耀[1] 

机构地区：[1]湖南软件职业学院软件工程系,湘潭411100

出　　处：《计算机系统应用》2012年第10期152-154,203,共4页Computer Systems & Applications

基　　金：湖南省教育厅科学研究项目(11C0724;11C0723)

摘　　要：由于电信市场竞争日益激烈,为了保持客户,防止客户流失,提升企业的竞争力,利用DBSCAN算法对流失客户群数据进行划分,分析客户流失原因,结合决策树ID3算法进行客户流失预测,实验结果表明,两种算法的结合,使得客户流失预测准确率得到较大提高.Due to an increasingly competitive telecommunications market, in order to Keep customers, to prevent me loss of customers, enhance the competitiveness of enterprises, this paper uses DBSCAN algorithm to the loss of customers data types, analysis of customer loss, combined with the decision tree ID3algorithm in customer churn prediction, experimental results show that, two kinds of algorithm combined, makes the customer churn prediction accuracy obtained improved.

关 键 词：数据挖掘 DBSCAN算法 决策树ID3算法 客户保持 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论]