期刊名称：International Journal of Innovative Research in Computer and Communication Engineering
印刷版ISSN：2320-9798
电子版ISSN：2320-9801
出版年度：2013
卷号：1
期号：2
出版社：S&S Publications
摘要：Now a day the usage of credit cards has dramatically increased. As credit card becomes the most popularmode of payment for both online as well as regular purchase, cases of fraud associated with it are also rising. Varioustechniques like classification, clustering and apriori of web mining will be integrated to represent the sequence ofoperations in credit card transaction processing and show how it can be used for the detection of frauds. Initially, webmining techniques trained with the normal behaviour of a cardholder. If an incoming credit card transaction is notaccepted by the web mining model with sufficiently high probability, it is considered to be fraudulent. At the sametime, the system will try to ensure that genuine transactions will not be rejected. Using data from a credit card issuer, aweb mining model based fraud detection system will be trained on a large sample of labelled credit card accounttransactions and tested on a holdout data set that consisted of all account activity. Web mining techniques can betrained on examples of fraud due to lost cards, stolen cards, application fraud, counterfeit fraud, and mail-order fraud.The proposed system will be able to detect frauds by considering a cardholder‟s spending habit without its significance.Usually, the details of items purchased in individual transactions are not known to any Fraud Detection System. Theproposed system will be an ideal choice for addressing this problem of current fraud detection system. Anotherimportant advantage of proposed system will be a drastic reduction in the number of False Positives transactions. FDSmodule of proposed system will receive the card details and the value of purchase to verify, whether the transaction isgenuine or not. If the Fraud Detection System module will confirm the transaction to be of fraud, it will raise an alarm,and the transaction will be declined.
