Soft computing: Investigation of two-step approach to identity theft detection in electronic payments

Identity theft and identity fraud are terms used to refer to all types of crime in which someone wrongfully obtains and uses another person's personal data in some way that involves fraud or deception, typically for economic gain. In e-commerce, the growing number of credit card transactions provides more opportunity for thieves to steal credit card numbers and subsequently commit fraud. Despite significant efforts by merchants, card issuers and law enforcement to curb fraud, online fraud continues to plague electronic commerce web sites. In this research work, Communal detection (CD) and Counter Propagation Neural Network (CPNN) were employed for fraud detection in identity theft in credit cards online transactions. Three thousand credit cards transactions were simulated. The selected simulated applicants attributes were worked on by CD to produce whitelists and blacklists. The result-cum-applicants data were preprocessed and passed to the CPNN to performed classifications. The results showed that CD-CPNN system produced average false positive rate, average false alarm rate, average detection rate and average prediction accuracy of 26.3, 2.7, 93.6, and 92.5%; CD system generated 67.2, 6.8, 90.4 and 85.3%; while CPNN system gave 64.8, 7.4, 90.5 and 88.7% respectively.