PERFORMANCE EVALUATION OF UNSUPERVISED NEURAL NETWORK IN FRAUD DETECTION
*Ismaila W.O. 1 Alese B. K.2 Adeosun O. O.1 Arulogun O. T. 1
1Ladoke Akintola University of Technology, Ogbomoso.
2Federal University of Technology, Akure.
ABSTRACT
Despite significant efforts by merchants, card issuers and law enforcement to curb fraud, online fraud continues to plague electronic commerce web sites. More advanced solutions are desired to protect merchants from the constantly evolving problem caused by fraud. The supervised machine learning technique for the most well known fraud detection algorithms makes them inadequate for an online system. This paper presents an automated credit card fraud detection system based on the unsupervised neural network technology. The proposed system is based on Self-Organizing Map algorithm that creates a model of typical cardholder’s spending profiles to detect suspicious transactions. The results were evaluated with performance metrics to determine its effectiveness.
Keywords: Payment System, Credit card, Spending profiles, Fraud Detection, Self Organizing Map.