Analysis of Deep Learning Techniques for Bank Fraud Detection

The banking industry significantly contributes to a nation's development and economic improvement, but the rise of smart transactions in recent times where customers conduct trade both online and physically, has raised concerns about fraud. The availability of advanced technology has made online banking a prominent channel for business but it has also resulted in a surge of fake banking activities and fraudulent transactions compromising the security and trust of users. These fraudulent activities such as hacking user information, spreading viruses and scams, and creating fake links have resulted in significant losses. The financial institutions have placed various methods including artificial intelligence and machine learning based solutions for early fraud detection. However, implementation of these solutions still remains a challenge due to the practical problems associated with these transactions, as data set is highly imbalanced, there is lack of sufficient data and large stake in false negative rates. Majority of the machine learning models trained on such data sets perform well on accuracy-based performance criteria however, do not produce good ROC-AUC performance due to excessive false positive or false negative rates. The CNN model is proposed to get better performance metrics.