Analysing Machine Learning Algorithms for Identifying Fraudulent Financial Transactions

Abstract

As we witnessthe era of digital payments and trans- actions, which has increased exponentially because of the ever- emerging advanced technology, recognizing fraudulent and legitimate transactions has become a volatile challenge that banking institutions face these days. The resulting by-product of digital transformation is a significant rise in fraudulent activities that affect banking institutions, merchants, and banks. Traditional methods of identifying such statements, including manual auditing and inspection, are costly, imprecise, and time-consuming. Machine learning algorithms and other latest technologies are being applied in the financial sector to support trading activities, mobile banking, payments, and making customer credit decisions. In this comprehensive study, I have reviewed many methods of machine learning that detect financial fraud. Since Ensemble Learning techniques were employed more than unsupervised and supervised methods like clustering and random forest. The next research and analysis should intensify the focus on unsupervised semi-supervised for fraud detection in new emerging problems in the field of digital financial fraud.

Keywords: Machine Learning, Digital Payments, Financial Fraud, Ensemble Learning, Supervised Learning