管理学报
  125年5月29日 星期四
首页 |  期刊介绍 |  编 委 会 |  投稿须知 |  期刊订阅 |  广告服务 |  联系我们 |  留言板 | English
J4  2012, Vol. 9 Issue (9): 1373-    DOI:
营销与服务 最新目录| 下期目录| 过刊浏览| 高级检索 |
基于多分类器动态选择与成本敏感优化集成的 电信客户流失预测研究
罗彬, 邵培基, 夏国恩
1. 电子科技大学经济与管理学院; 2. 广西财经学院工商管理系
A Study on Prediction of Telecom Customer Churn Based on Dynamic Selection, Optimization and Integration of Cost Sensitivity
LUO Ban, SHAO Pei-Ji, XIA Guo-En
1. University of Electronic Science and Technology of China, Chengdu, China; 2. Guangxi University of Finance and Economics, Nanning, China

全文: PDF (185 KB)   HTML (1 KB) 
输出: BibTeX | EndNote (RIS)      
摘要 针对不同样本在特征空间中具有不同的区域特性和不同分类算法之间的预测互补性,在电信客户流失预测理论基础上,融合多分类器动态集成理论和成本敏感学习理论,建立了电信客户流失多分类器集成预测的利润函数,并提出了一类新的基于多分类器动态选择与成本敏感优化集成的电信客户流失预测模型。首先使用K均值聚类法聚类训练样本成多个分区;接着使用NaiveBayes算法、多层感知机算法和J48算法在各分区样本上构建客户流失预测子分类器;最后使用改进人工鱼群算法分别对各分区的子分类器进行成本敏感优化集成。实验结果表明,所提出的基于多分类器动态选择与成本敏感优化集成模型的分类性能不仅优于由训练集全体样本所构建的3个单模型,也优于基于改进人工鱼群算法优化集成这3个单模型而得到的集成模型。
服务
把本文推荐给朋友
加入我的书架
加入引用管理器
E-mail Alert
RSS
作者相关文章
罗彬
邵培基
夏国恩
关键词 客户流失预测多分类器动态选择成本敏感优化集成成本敏感学习人工鱼群算法    
Abstract:On account that the different samples have the prediction complementarities between different section characters and different classification algorithms in feature space and based on the theory of Telecom customer churn prediction, this paper established the profits functions to predict Telecom customer churn integrating multiclassifiers, and a new customer churn prediction model is put forward in Telecom based on the dynamic selection and optimizing integrating of cost sensitivity. Firstly, the training set samples are clustered into multiple subareas by using Kmeans clustering algorithm. Then, the customer churn prediction subclassifiers are established based on the samples in the subareas by using NaiveBayes Algorithm, Multilayer Perceptron and J48 Algorithm, respectively. Finally, the subarea subclassifiers are integrated and optimized by use of the Improved Artificial Fishschool Algorithm(IAFSA). The experiment results show that the classifying performance of the model based on the dynamic integration of multiclassifiers and optimizing integrating of cost sensitivity not only excels the three single model constructed based on the whole samples, but also excels the model integrating of the three single model by IAFSA.
Key wordscustomer churn prediction    dynamic selection of multiple classifiers    optimizing integration of cost sensitivity    cost sensitivity learning    artificial fishschool algorithm (AFSA)   
收稿日期: 2010-01-29     
基金资助:国家自然科学基金资助项目(70801021);中国博士后科学基金资助项目(20080431276);教育部人文社会科学资助项目(08JC630019)
通讯作者: 罗彬(1974~),男,四川渠县人。电子科技大学经济与管理学院(成都市610054)博士研究生。研究方向为商务智能研究。     E-mail: Luobin10000@163.com
引用本文:   
罗彬, 邵培基, 夏国恩. 基于多分类器动态选择与成本敏感优化集成的 电信客户流失预测研究[J]. J4, 2012, 9(9): 1373-. LUO Ban, SHAO Pei-Ji, XIA Guo-En. A Study on Prediction of Telecom Customer Churn Based on Dynamic Selection, Optimization and Integration of Cost Sensitivity. J4, 2012, 9(9): 1373-.
链接本文:  
http://manu68.magtech.com.cn/Jwk_glxb/CN/     或     http://manu68.magtech.com.cn/Jwk_glxb/CN/Y2012/V9/I9/1373
版权所有 © 《管理学报》编辑部
本系统由北京玛格泰克科技发展有限公司设计开发  技术支持:support@magtech.com.cn