基于特征选择的存量客户流失预警分析  

Early Warning Analysis of Stock Customer Churn Based on Feature Selection

作　　者：李琼阳 何月华 LI Qiongyang;HE Yuehua(School of Science,Xuchang University,Xuchang 461000,China)

机构地区：[1]许昌学院数理学院,河南许昌461000

出　　处：《许昌学院学报》2022年第5期12-15,共4页Journal of Xuchang University

基　　金：国家自然科学基金青年项目(项目编号12001471)。

摘　　要：随着携号转网政策的施行,电信运营商面临着客户流失的困境,为了高效地识别出流失风险较高的客户,首先计算了各指标的信息值,筛选出来重要的建模特征,继而利用这些特征构建决策树模型,将现存客户进行分类.最后通过实证分析,并且将决策树模型与逻辑回归模型的分类结果进行对比,验证了决策树模型的有效性和实用性.With the implementation of the number transfer policy,telecom operators are faced with the dilemma of customer loss.In order to efficiently identify customers with high risk of customer loss,the information values of each indicator are firstly calculated,and important modeling features are selected.Then,the decision tree model is constructed by using these features to classify existing customers.Finally,the validity and practicability of decision tree model are verified by empirical analysis and comparison of classification results between decision tree model and logistic regression model.

关 键 词：携号转网 流失预警 信息值 决策树 

分 类 号：TP391.7[自动化与计算机技术—计算机应用技术]