Online Transaction Fraud Detection and Prevention Using Hmm and Behavior Analysis

Abstract

The objective of this project is to both prevent and detect fraudulent transactions. As the use of online transactions, such as NEFT and RTGS, has grown with the rapid advancement of internet technology, credit card payments have become a popular method of payment for both offline and online purchases. However, this has also led to an increase in online transaction fraud. Criminals have taken advantage of the popularity of network transactions to commit crimes, causing banks to suffer significant losses. The current fraud detection system only identifies fraudulent transactions after they have been completed. To address this issue, we will use the Hidden Markov Model (HMM). This will allow for fraud to be detected during the transaction and immediately blocked by sending a verification code to the user's email address. To improve the system's effectiveness, we have incorporated a real-time payment gateway method. Furthermore, we will utilize behavior analysis to comprehend the user's spending habits. This combination of HMM and behavior analysis will facilitate advanced fraud analysis with a low false alarm ratio.