电信科学 ›› 2009, Vol. 25 ›› Issue (3): 70-75.doi: 10.3969/j.issn.1000-0801.2009.03.021

• 研究与开发 • 上一篇    下一篇

基于贝叶斯网络的移动业务客户流失预测研究

桂宏新1,杨昌昊2,程飞3   

  1. 1 中国联通集团有限公司合肥分公司 合肥 230009
    2 中国科学技术大学 合肥 230027
    3 合肥工业大学计算机网络所 合肥 230009
  • 出版日期:2009-03-15 发布日期:2009-03-15
  • 基金资助:
    国家自然科学基金资助项目

Research on Customer Loss Prediction for Mobile Service Based on Bayesian Network

Hongxin Gui1,Changhao Yang2,Fei Cheng3   

  1. 1 Hefei Branch of China Unicom Limited Company，Hefei 230009，China
    2 University of Science and Technology of China，Hefei 230027，China
    3 Institute of Computer Network System，Hefei University of Technology，Hefei 230009，China
  • Online:2009-03-15 Published:2009-03-15

在线阅读

17

PDF下载

1093

被引次数

16

摘要：

准确的客户流失预测能够使移动业务针对特定用户需求和消费水平进行差异化服务。本文提出了一种基于贝叶斯网络的客户流失预测模型，根据案例分析的结果识别客户离网倾向原因，寻找具有高离网倾向的客户群，通过分类的特征分析，得出客户流失规则，并验证模型的有效性。

关键词: 客户流失, 贝叶斯网络, 移动业务, 预测

Abstract:

Accurate prediction of customers loss could enable mobile business to offer different service for specific needs and consumption of users. A model for the forecast of the loss of customers had been constructed based on Bayesian network. And according to the results of case studies，the paper identified the reasons for customer churn tend to look for a customer base with a high tendency to off-grid，and through the characteristics analysis of the classification，the rules of loss of customers had been arrived at and the effectiveness of the model had been verified.

Key words: customer loss, Bayesian network, mobile service, prediction

No Suggested Reading articles found!