Building the mobile VIP Customer Churn Prediction Model
VIP customers are very important wealth of mobile operators.VIP customers churn management is very critical to gain competition advantage.Prediction who will churn is a major step in the churn management process.One of the challenges is that churner prediction specific to VIP is not available in existing literatures,especially on how to select predictive variables and deal with skewing data.In the study,we analyze the features of VIP customers,and compare various techniques to solve data skewing.The results indicate that our models can achieve satisfactory prediction effectiveness by using customer static and dynamic attributes,and improve the prediction performance on skewing data.
churn management VIP customers decision tree skewing data
Qian Su Peiji Shao Tao Zou
School of economic & management,UESTC Chengdu,China Department of data analysis,WXTK Technology CO.Chengdu,China
国际会议
太原
英文
328-331
2011-02-26（万方平台首次上网日期，不代表论文的发表时间）