Fraud Detection in Mobile Communications Using Rule-Based and Neural Network System

Fraud detection involves identifying fraud as quickly as possible once it occurs. The increased use of mobile communications has resulted in the collection of large amounts of data. Information and knowledge obtained from these databases can give operators a competitive edge in terms of fraud detection. A fraud detection tool utilizing both rule-based and neural network technologies that enable the profiling of network subscribers and network traffic is developed. The performance evaluation showing the percentage of correctly identified fraudsters versus the percentage of new subscribers raising alarms are optimized using the Receiver-Operating Characteristic (ROC) curve for both the training and test sets, resulting in very low false positive rates