Realistic Synthetic Financial Transactions for Anti-Money Laundering Models

TL;DR

The paper addresses the scarcity and labeling challenges of real AML data by introducing AMLworld, a multi-agent synthetic data generator that produces richly labeled financial transactions, including laundering events. It provides a suite of realistic AML datasets (HI/LI, Small/Medium/Large) publicly on Kaggle, enabling robust benchmarking and cross-bank experiments. The approach models a full money-laundering lifecycle within a virtual world, including shell entities and multi-currency transactions, and validates the utility of both Graph Neural Networks and gradient-boosted trees for detection, with transfer learning and data sharing offering notable gains. This synthetic benchmark supports what-if analyses, federated learning considerations, and ethical use for improving AML defenses without exposing real customer data. The work thus offers a standardized, scalable platform for developing and evaluating AML models in a realistic, controllable setting with complete ground truth.

Abstract

With the widespread digitization of finance and the increasing popularity of cryptocurrencies, the sophistication of fraud schemes devised by cybercriminals is growing. Money laundering -- the movement of illicit funds to conceal their origins -- can cross bank and national boundaries, producing complex transaction patterns. The UN estimates 2-5\% of global GDP or \$0.8 - \$2.0 trillion dollars are laundered globally each year. Unfortunately, real data to train machine learning models to detect laundering is generally not available, and previous synthetic data generators have had significant shortcomings. A realistic, standardized, publicly-available benchmark is needed for comparing models and for the advancement of the area. To this end, this paper contributes a synthetic financial transaction dataset generator and a set of synthetically generated AML (Anti-Money Laundering) datasets. We have calibrated this agent-based generator to match real transactions as closely as possible and made the datasets public. We describe the generator in detail and demonstrate how the datasets generated can help compare different machine learning models in terms of their AML abilities. In a key way, using synthetic data in these comparisons can be even better than using real data: the ground truth labels are complete, whilst many laundering transactions in real data are never detected.

Figures (11)

• Figure 1: Financial transactions in (a) tabular format and in (b) graph format.
• Figure 2: Laundering Patterns Modelled
• Figure 3: Overview of AMLSim.
• Figure 4: Overview of AMLworld Simulation.
• Figure 5: Illustration of the entity / circular flow graph representing our Virtual World Model.