Customer Churn Prediction of China Telecom Based on Cluster Analysis and Decision Tree Algorithm

Abstract

In this paper, cluster analysis is applied in the customer data provided by a branch of China Unicom in Guangdong province. It is established that customer churn prediction model of China Telecom based on cluster analysis and decision tree algorithm. The prediction model can provide scientific basis and reference for maintaining and retaining customers of China Telecom, since it can efficiently discover the valuable customers with leaving orientation from massive information of customers. The experimental results show that the customer churn prediction model in this paper is effective. © Springer-Verlag Berlin Heidelberg 2012.