Identifying fake and potential corporate members in telecommunications operators

Nowadays, mobile operators in China mainland are facing fierce competition from one to another, and their focus of customer competition has, in general, shifted from public to corporate customers. One big challenge in corporate customer management is how to identify fake corporate members and potential corporate members from corporate customers. In this study, we have proposed an identification method that combines the rule-based and probabilistic methods. Through this method, fake corporate members can be eliminated and external potential members can be mined. The experimental results based on the data obtained from a local mobile operator revealed that the proposed method can effectively and efficiently identify fake and potential corporate members. The proposed method can be used to improve the management of corporate customers.