Internet Banking Fraud Detection Using Deep Learning Based on Decision Tree and Multilayer Perceptron

Fraud transactions have become a growing problem in the online banking sphere. As technology progresses, fraudsters also change their methods of committing fraud. There are also emerging technologies that allow fraudsters to mimic the transaction behavior of genuine customers and they also keep changing their methods so that it is difficult to detect fraud. This paper discusses the importance of fraud detection methods and compares Hidden Markov Model, Deep Learning, and Neural Network that are used to detect fraud in online banking transactions.

I. Introduction

Online shopping systems are highly susceptible to fraud and detecting any such fraud has become top priority of online business owners. Fraud instances develop according to the system checks that are put in place to prevent them. Modern fraud investigators, banks and electronic payment systems work hard to detect fraud in early stages and prevent it from harming the customers’ accounts. CyberSource mentioned in a 2017 report that their web store faced fraud loss up to 74% and their mobile channels faced up to 49% fraud through order channel [8]. This information sets the groundwork for the changes in fraud behavior in the recent years with new advances in technology.