With the further reform and market division in the telecommunication industry, there are more and more choices for customers to select telecom products and operators, which lead to the fiercer competition for customers between telecom operators. As the technical method to identify customers churn, the data mining can help the telecom competitors to analyze some seemingly unrelated data into meaningful information. On the basis of the research on the vital problems in the telecom companies, this paper explains how to apply data mining techniques to customer churn analysis, proposes the specific procedures and technology solutions to prevent the customer churn and builds the models of the data mining by analyzing the related algorithm. Finally, based on the systematical analysis on theory and method to data mining, the paper draws the conclusion that the customers churn listing and tree algorithm can solve the practical problems of the customer churn in telecommunication industry.
One-Step Dynamic Classifier Ensemble Model for Customer Value Segmentation with Missing Values