Method and System for Detecting Fraud in Credit Card Transaction

Due to a rapid advancement in the electronic commerce technology. Credit card becomes the most popular mode of payment for both online as well as regular purchase. Cases of fraud associated with it are also rising. In this paper I am introducing the concept of three level of security, the first level is the static User name or password, and in the second level it uses Hidden Markov Model (HMM) and shows how it can be used for the detection of frauds. An HMM is initially trained with the normal behaviour of a cardholder. If an incoming credit card transaction is not accepted by the trained HMM with sufficiently high probability, it is considered to be fraudulent. At the same time, we try to ensure that genuine transactions are not rejected. And to reduce the false positive transactions we will send the dynamic password, which can be send through the use of web services to the user’s mobile phone number instantly and he/she has to enter same password for getting the authorization from the bank side and suppose if due to the heavy load on the server side , if the user does not get the password in its mobile phone within the given stipulated time , then after a little time interval some personnel questions(either security question or images) will be asked which can be answered by the end user