Review of Fraud Detection and Churn Behavior Modeling Techniques

Abstract

This research pursues inductive/deductive approach by studying existing techniques for fraud detection and customer churn prediction. Telecommunication operators for instance store large amounts of data related with the activity of their clients. In these records exists both normal and fraudulent activity records. It is expected that the fraudulent activity records should be substantially smaller than the normal activity. If it were the other way around this type of business would be impractical due to the amount of revenue lost. Furthermore, customer churn is the focal concern of most companies which are active in industries with low switching cost. Among all industries which suffer from this issue, telecommunications industry can be considered in the top of the list with approximate annual churn rate of 30%. This means wasting the money and efforts, "it is like adding water to a leaking bucket". In this paper, the authors present the review of past work that has been carried out by various researchers based on fraud detection and churn behavior modeling.