Hidden Markov Model for Online Credit Card Transaction Fraud Detection

Abstract

Now a days,Credit card fraud is a wide-ranging term for theft and fraud committed using or involving a payment card, such as a credit card as a fraudulent source of funds in a transaction. The most accept mode is credit card for both online and offline in today's world. It provide cashless shopping at every shop in all countries. So as credit card is becoming most popular mode for online financial transactions, at the same time fraud associated with it are also rising. In this paper HMM(Hidden Markov Model) is used to model sequence of operation in credit card transaction processing. HMM does not required fraud signatures it is initially trained with the normal behavior of a cardholder. If an incoming Online card transaction is not accepted by the trained HMM with sufficiently high probability, it is considered to be fraudulent. At the same time, we will attempt to guarantee that fraud transactions are rejected.At the same time, we will attempt to guarantee that fraud transactions are rejected.