Credit Card Customer Churn Prediction Based on the RST and LS-SVM

摘要：

The credit card business in the bank possesses high risk and high profit. How to control the customer churn of credit card has already become the problem to solve in the urgent need. In order to support the bank to reduce churn rate, we need to predict which customers are high risk of churn and optimize their marketing intervention resource to prevent as many customers as possible from churning. Considering the shortcomings of conventional prediction methods, Rough Set Theory (RST) and Least Squares Support Vector Machine (LS-SVM) is adopted to establish the prediction model of credit card customer churn, which could predict the customer churn efficiently and effectively. Predicting the tendency of customer churn according to LS-SVM will provide a scientific guide for the credit card customer marketing of the bank.

关键词： Credit Card Customer Churn prediction RST LS-SVM

作者: Ning Wang Dong-xiao Niu

作者单位: School of Business Administration,North China Electric Power University,and Beijing

会议类型: 国际会议

会议名称: 2009 6th International Conference on Service Systems and Service Management( 2009 第六届服务系统与服务管理国际会议)

会议地点: 厦门

会议语种:英文

页码: 275-279

在线出版日期: 2009-06-08（万方平台首次上网日期，不代表论文的发表时间）