Shuvo, Showive Hasan and Rahman, Jarin Nuran and Amin, Al and Shurovi, Shoumya Hassan (2026) A Systematic Review of Machine Learning and Graph-Based Approaches for Financial Fraud Detection. American Journal of Economics and Business Management, 9 (10). ISSN 2576-5973

Text AJEBM_A+Systematic+Review.pdf Download (579kB)

Abstract

The growing complexities of digital transactions and the rise in digital fraud pose a significant threat to financial transactions. In this research, a systematic review is presented on financial fraud detection using machine learning and graph-based methods. The survey discusses the related works of the AI techniques, feature engineering techniques, behavior analysis and graph based fraud detection techniques. Financial transaction data is analyzed using exploratory data analysis to detect fraud patterns in terms of geographical location, amount of financial transaction, risk score and the distribution of financial transactions. The results show that there are salient behavioral characteristics and risk related attributes that offer important clues in detecting suspicious activities. The study emphasizes the benefits of machine learning methods for better fraud detection in automated systems, as they can be used to learn from previous transactions, and of graph-based methods as they can be used to identify relationships between financial entities. The study offers an overview of the potential benefits of combining machine learning with graph-based techniques for financial fraud detection, as well as the potential drawbacks of traditional methods.

Item Type:	Article
Uncontrolled Keywords:	Digital transaction, AI techniques, Financial Transaction, Machine Learning Methods, Graph-based Fraud Detection
Subjects:	H Social Sciences
Depositing User:	admin eprints
Date Deposited:	22 Oct 2026 02:47
Last Modified:	22 Oct 2026 02:47
URI:	http://eprints.umsida.ac.id/id/eprint/17262

Actions (login required)

View Item