Fraud Detection in Telecommunications Industry: Bridging the Gap with Random Rough Subspace Based Neural Network Ensemble Method

Abstract

Fraud has been very common in the society and it affects private enterprises as well as public entities. Telecommunication companies worldwide suffer from customers who use the provided services without paying. There are also different types of telecommunication fraud such as subscription fraud, clip on fraud, call forwarding, cloning fraud, roaming fraud and calling card fraud. Thus, detection and prevention of these frauds are the main targets of the telecommunication industry. This paper addresses the various techniques of detecting fraud, giving the limitations of each technique and proposes random rough subspace-based neural network ensemble method for effective fraud detection.

Keywords: Fraud, Fraud detection, Random rough subspace, Neural network, Telecommunications