XDHG-AML: Intrinsically Explainable Dynamic Heterogeneous Graph Learning for Anti-Money Laundering Detection
Keywords:
anti-money laundering, dynamic graphs, heterogeneous graphs, explainable graph learning, fraud detectionAbstract
Anti-money laundering (AML) detection is naturally a dynamic heterogeneous graph problem: accounts exchange funds over time, share authentication devices, and interact with merchants or counterparties, while analysts require evidence that connects a risk score to concrete relational behavior. This paper proposes XDHG-AML, an intrinsically explainable dynamic heterogeneous graph model that produces a laundering score together with a normalized relation-time evidence map. The model combines relation-specific message aggregation, a learnable relation evidence gate, a gated recurrent temporal encoder, and a time evidence gate. A sparsity-and-smoothness regularizer concentrates explanations on a small set of relation-time cells without introducing a separate post-hoc explainer. We evaluate the method on a controlled dynamic AML benchmark generated from four laundering typologies—fan-in, fan-out, circular flow, and shared-device coordination—over 2,200 accounts and 10 temporal snapshots. Across three independently initialized runs, XDHG-AML obtains 0.916±0.060 ROC-AUC, 0.578±0.176 average precision, and 0.545±0.113 F1. Relative to a static heterogeneous model, dynamic modeling improves ROC-AUC by 5.2 percentage points and F1 by 10.8 points. Masking the highest-evidence relation-time cell lowers the laundering probability of positive test cases by 3.59 percentage points on average, whereas random masking changes it by -0.12 points, providing an intervention-based fidelity check. These results show that dynamic heterogeneous modeling can materially improve AML discrimination while yielding compact evidence aligned with the model computation.
Index Terms—
References
[1] M. Weber, G. Domeniconi, J. Chen, D. K. I. Weidele, C. Bellei, T. Robinson, and C. E. Leiserson, “Anti-money laundering in Bitcoin: Experimenting with graph convolutional networks for financial forensics,” KDD 2019 Workshop on Anomaly Detection in Finance, 2019.
[2] Fang, Q., & Liu, W. (2025). HARLA-ED: Resolving Information Asymmetry and Enhancing Algorithmic Symmetry in Intelligent Educational Assessment via Hybrid Reinforcement Learning. Symmetry, 18(1), 58.
[3] Mai, N. T., Cao, W., & Fang, Q. (2025). A study on how LLMs (eg GPT-4, chatbots) are being integrated to support tutoring, essay feedback and content generation. Journal of Computing and Electronic Information Management, 18(3), 43-52.
[4] E. Altman, J. Blanuša, L. von Niederhäusern, B. Egressy, A. Anghel, and K. Atasu, “Realistic synthetic financial transactions for anti-money laundering models,” Advances in Neural Information Processing Systems, vol. 36, Datasets and Benchmarks Track, 2023.
[5] T. N. Kipf and M. Welling, “Semi-supervised classification with graph convolutional networks,” International Conference on Learning Representations, 2017.
[6] M. Schlichtkrull et al., “Modeling relational data with graph convolutional networks,” European Semantic Web Conference, 2018.
[7] X. Wang et al., “Heterogeneous graph attention network,” The World Wide Web Conference, pp. 2022–2032, 2019, doi: 10.1145/3308558.3313562.
[8] Z. Hu, Y. Dong, K. Wang, and Y. Sun, “Heterogeneous graph transformer,” The Web Conference, 2020.
[9] A. Pareja et al., “EvolveGCN: Evolving graph convolutional networks for dynamic graphs,” AAAI Conference on Artificial Intelligence, vol. 34, no. 4, pp. 5363–5370, 2020.
[10] D. Xu, C. Ruan, E. Korpeoglu, S. Kumar, and K. Achan, “Inductive representation learning on temporal graphs,” International Conference on Learning Representations, 2020.
[11] E. Rossi, B. Chamberlain, F. Frasca, D. Eynard, F. Monti, and M. Bronstein, “Temporal graph networks for deep learning on dynamic graphs,” arXiv:2006.10637, 2020.
[12] A. Sankar, Y. Wu, L. Gou, W. Zhang, and H. Yang, “DySAT: Deep neural representation learning on dynamic graphs via self-attention networks,” ACM International Conference on Web Search and Data Mining, 2020.
[13] Y. You, T. Chen, Y. Sui, T. Chen, Z. Wang, and Y. Shen, “Graph contrastive learning with augmentations,” Advances in Neural Information Processing Systems, vol. 33, 2020.
[14] Y. Jiao, H. Fan, X. Yue, W. Ping, T. Sun, and J. Wang, “Dynamic heterogeneous graph contrastive learning for uncovering collusive financial fraud,” Scientific Reports, vol. 16, Art. no. 28993, 2026, doi: 10.1038/s41598-026-58938-5.
[15] P. Veličković, W. Fedus, W. L. Hamilton, P. Liò, Y. Bengio, and R. D. Hjelm, “Deep graph infomax,” International Conference on Learning Representations, 2019.
[16] K. Hassani and A. H. Khasahmadi, “Contrastive multi-view representation learning on graphs,” International Conference on Machine Learning, pp. 4116–4126, 2020.
[17] Z. Ying, D. Bourgeois, J. You, M. Zitnik, and J. Leskovec, “GNNExplainer: Generating explanations for graph neural networks,” Advances in Neural Information Processing Systems, vol. 32, 2019.
[18] S. M. Lundberg and S.-I. Lee, “A unified approach to interpreting model predictions,” Advances in Neural Information Processing Systems, vol. 30, 2017.
[19] T.-Y. Lin, P. Goyal, R. Girshick, K. He, and P. Dollár, “Focal loss for dense object detection,” IEEE International Conference on Computer Vision, pp. 2980–2988, 2017, doi: 10.1109/ICCV.2017.324.
[20] K. Cho et al., “Learning phrase representations using RNN encoder-decoder for statistical machine translation,” Conference on Empirical Methods in Natural Language Processing, 2014.
[21] Wang, Y., & Xing, S. (2025). AI-driven CPU resource management in cloud operating systems. Journal of Computer and Communications, 13(06), 135-149.
[22] Xing, S., & Wang, Y. (2025). Cross-modal attention networks for multi-modal anomaly detection in system software. IEEE Open Journal of the Computer Society.
[23] Chen, T., Shang, W., & Guo, Z. (2024). Xsla: An SLA-aware heterogeneous XPU scheduling framework for edge AI inference. World Journal of Information Technology, 2(2), 48–59.
[24] Qin, Y., Zou, J., & Rhee, M. (2022). An AI-driven intelligent transaction execution framework for internet finance platforms: Cost-sensitive real-time integrity interdiction. Eurasia Journal of Science and Technology, 4(2), 36–44.
[25] Wang, Y., & Zhang, C. (2023). AERIS: Adaptive executor scoped resource management for privacy preserving data processing in Apache Spark. Eurasia Journal of Science and Technology, 5(2), 25-33.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Software Engineering and Artificial Intelligence Systems
This work is licensed under a Creative Commons Attribution 4.0 International License.
