Exploring Models with Social Network Analytics to Improve Churn Prediction of Mobile Phone Operators.

摘要

The research topic for this study was churn prediction models for post-paid and pre-paid mobile phone customers. The problem addressed in this study was the lack of models for predicting customer churn with the inclusion of social network factors. The limited effectiveness of current prediction models results in an average churn rate for mobile phone operators today that exceeds 20 percent, driving the global cost of customer defection to over