2026/01/01 by Dang Sy Duy, Sy-Duy Dang, Nguyen Duy Chien +3
Computer Science · #Advanced Graph Neural Networks #Anomaly detection #Artificial neural network #Blockchain Technology Applications and Security #Class (philosophy) #Database transaction #Financial fraud #Graph #Imbalanced Data Classification Techniques #Node (physics)
paper · pdf · doi:10.1007/978-3-032-31141-2_22
published in Communications in computer and information science, 313-326 (Springer Science+Business Media)
openalex publication_date 2026/01/01 · openalex created_date 2026/07/10 · openalex updated_date 2026/08/05
Graph neural networks (GNNs) offer a principled approach to financial fraud detection by jointly learning from node features and transaction graph topology. However, their effectiveness on real-world anti-money laundering (AML) benchmarks depends critically on training practices such as specifically weight initialisation and normalisation that remain underexplored. We present a systematic ablation of initialisation and normalisation strategies across three GNN architectures (GCN, GAT, and GraphSAGE) on the Elliptic Bitcoin dataset. Our experiments reveal that initialisation and normalisation are architecture-dependent: GraphSAGE achieves the strongest performance with Xavier initialisation alone, GAT benefits most from combining GraphNorm with Xavier initialisation, while GCN shows limited sensitivity to these modifications. These findings offer practical, architecture-specific guidance for deploying GNNs in AML pipelines for datasets with severe class imbalance. We release a reproducible experimental framework with temporal data splits, seeded runs, and full ablation results.