From Transactions to Networks: Leveraging Graph Algorithms for Money Laundering Detection in Banking Systems

Money laundering poses a significant threat to the integrity of financial systems, enabling criminal activities by disguising illicit funds as legitimate transactions. Detecting these schemes within banking networks is challenging due to their complex, dynamic, and often obfuscated nature. Recent advances in graph-based analytics offer promising solutions by modeling financial transactions as networks, where nodes represent entities (e.g., accounts, customers) and edges denote monetary flows. This paper provides a comprehensive review of graph algorithms applied to uncovering money laundering networks in banking systems. We examine key techniques such as community detection, centrality analysis, anomaly detection, and subgraph pattern matching, highlighting their effectiveness in identifying suspicious behaviors. Additionally, we discuss real-world case studies, computational challenges, and future directions for enhancing detection accuracy using machine learning and scalable graph processing frameworks. Our review underscores the potential of graph-based approaches in strengthening anti-money laundering (AML) efforts while addressing limitations such as false positives and evolving laundering tactics.

کلیدواژه ها:

Money laundering detection ، graph algorithms ، financial networks ، anomaly detection ، anti-money laundering (AML)