Skip to Main Content
Telecommunications fraud is increasing dramatically resulting in costing operators almost $60 billion each year. Due to the increase of bandwidth of 3G mobile communication equipment, the fraud can be in terms of network, commercial, customer or even staff and all will remain keys in assessing 3G fraud risk. Current methods such as artificial neural network and statistic methods used to detect fraud in 2.5G network canpsilat effectively extract intrinsic characteristics from huge database. This paper presents a novel rough fuzzy set based approach to detect fraud in 3G mobile telecommunication network. It analyzes the scenarios in 3G network including subscription fraud and superimposed fraud and profile and confirms the parameters to detect the scenarios. A rough fuzzy set based model to reduce the parameters in each scenario and get the refined characteristics. A rule based system called Citi FMS was designed to detect abnormities and alarm. The proposed system presented a framework used for 3G fraud risk including subscription and superimposed fraud.