基于营销策略的用户离网预测模型  

Customer Churn Prediction Model Based on Market Plan

作　　者：刘光远[1] 董立岩[2] 苑森淼[2] 任辉明[2] 

机构地区：[1]吉林大学通信工程学院,长春130025 [2]吉林大学计算机科学与技术学院,长春130012

出　　处：《吉林大学学报（信息科学版）》2007年第2期173-176,共4页Journal of Jilin University（Information Science Edition）

基　　金：国家自然科学基金资助项目(60275026)

摘　　要：传统的用户离网预测方法都是针对用户信息、消费行为等数据,通过数据挖掘的方法进行预测。为此,在分析移动用户离网原因的基础上提出了一种基于营销策略的用户离网预测模型(CPMP:Churn Predictionbased on Market Plan)。该模型针对用户主动离网的主要原因是其他运营商推出新的营销策略这一事实,通过比较用户在不同营销策略下可能发生的行为,进而预测用户离网的可能性。实验结果表明,基于CPMP模型对不同营销方案比较所得出差异,直接影响用户离网率。通过对营销策略的对比分析,可有效控制离网率。We bring forward a novel CPMP （Churn Prediction based on Market Plan） Model by studying the reasons of customer chum. Conventional customer churn prediction methods study customer＇s information and behavior, and use data mining method to predict. CPMP model aims at the most important fact that the reason of customer churn is from other companies＇market plans. Based on that fact, for each current customer, CPMP calculates the cost difference between self company and competitor companies, then predicts whether the customer will chum. The experiment result shows that the results compared by CPMP model of different market plans influence the customer churn rate. So by analysising different market plans the customer churn rate can be efficiently controlled.

关 键 词：数据挖掘 用户流失 营销策略 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]