LineMVGNN: Anti-Money Laundering with Line-Graph-Assisted Multi-View Graph Neural Networks

Researchers developed LineMVGNN, a novel graph neural network method for anti-money laundering that uses multi-view graph learning to analyze transaction networks. The method outperformed existing approaches on real-world datasets including Ethereum phishing networks and financial payment data.

• →LineMVGNN combines spatial graph neural networks with line-graph assistance to better detect money laundering patterns in transaction networks.
• →The method addresses limitations of conventional rule-based AML systems that rely heavily on domain knowledge and lack scalability.
• →Testing on Ethereum phishing networks and real financial payment data showed superior performance versus state-of-the-art methods.
• →The approach considers both payment and receipt transactions through two-way message passing between network nodes.
• →The research includes analysis of scalability, adversarial robustness, and regulatory considerations for practical deployment.