AMLNet: A Knowledge-Based Multi-Agent Framework to Generate and Detect Realistic Money Laundering Transactions
This addresses the problem of limited data for AML researchers and practitioners, offering a synthetic dataset and detection tool, though it is incremental as it builds on existing synthetic generation and detection methods.
The paper tackles the lack of publicly shareable, regulation-aligned transaction datasets for anti-money laundering (AML) research by introducing AMLNet, a knowledge-based multi-agent framework that generates 1,090,173 synthetic transactions with 0.16% laundering-positive cases and achieves 75% regulatory alignment and a 0.75 technical fidelity score, while its detection ensemble attains an F1 score of 0.90 on internal tests.
Anti-money laundering (AML) research is constrained by the lack of publicly shareable, regulation-aligned transaction datasets. We present AMLNet, a knowledge-based multi-agent framework with two coordinated units: a regulation-aware transaction generator and an ensemble detection pipeline. The generator produces 1,090,173 synthetic transactions (approximately 0.16\% laundering-positive) spanning core laundering phases (placement, layering, integration) and advanced typologies (e.g., structuring, adaptive threshold behavior). Regulatory alignment reaches 75\% based on AUSTRAC rule coverage (Section 4.2), while a composite technical fidelity score of 0.75 summarizes temporal, structural, and behavioral realism components (Section 4.4). The detection ensemble achieves F1 0.90 (precision 0.84, recall 0.97) on the internal test partitions of AMLNet and adapts to the external SynthAML dataset, indicating architectural generalizability across different synthetic generation paradigms. We provide multi-dimensional evaluation (regulatory, temporal, network, behavioral) and release the dataset (Version 1.0, https://doi.org/10.5281/zenodo.16736515), to advance reproducible and regulation-conscious AML experimentation.