GCF-MLD: Integrated Approach for Money Laundering Detection Using Machine Learning and Graph Network Analysis

Money laundering is a critical issue for financial institutions especially in developing countries and detecting such suspicious activities is a challenging task. The rule-based money laundering detection systems currently being used by the financial institutions rely on pre-defined rules that are often not able to keep up with rapidly changing tactics of money launderers. In this regard, machine learning based approaches have gained attention for detecting suspicious transactions related to money laundering. In this study, we propose a novel approach "Graph-Clustering Fusion for Money Laundering Detection (GCF-MLD)" combining graph network analysis with clustering methodology to identify suspicious accounts in real-world banking dataset. The results exhibit that the proposed model can be effectively used by the financial institutions to replace current rule-based techniques to significantly improve the efficiency and effectiveness of money laundering detection systems.