Related Experiment Video
Updated: Jul 15, 2025
10:17
High-throughput and Comprehensive Drug Surveillance Using Multisegment Injection-Capillary Electrophoresis-Mass Spectrometry
Published on: April 23, 2019
9.7K
A synthetic data set to benchmark anti-money laundering methods
Rasmus Ingemann Tuffveson Jensen1,2, Joras Ferwerda3, Kristian Sand Jørgensen4
1Department of Electrical and Computer Engineering, Aarhus University, Aarhus, 8200, Denmark. rasmus.tuffveson.jensen@ece.au.dk.
Scientific Data
|September 28, 2023
Summary
Researchers developed SynthAML, a synthetic dataset for anti-money laundering (AML) methods, addressing the lack of public bank data. This dataset enables robust benchmarking of AML techniques, showing real-world applicability.
Area of Science:
- Financial Criminology
- Data Science
- Machine Learning
Background:
- Publicly available bank transaction datasets are scarce due to confidentiality, hindering research in anti-money laundering (AML).
- This data scarcity limits investigations into critical AML challenges like efficiency, effectiveness, class imbalance, concept drift, and interpretability.
Purpose of the Study:
- To introduce SynthAML, a novel synthetic dataset designed for benchmarking statistical and machine learning methods in AML.
- To provide a realistic and accessible resource for advancing AML research and development.
Main Methods:
- SynthAML was generated using real data from Spar Nord, a Danish bank.
- The dataset comprises over 16 million transactions and 20,000 anti-money laundering alerts.
- It serves as a benchmark for evaluating various AML algorithms.
Main Results:
- Experimental results demonstrate that machine learning models trained on SynthAML exhibit transferable performance to real-world scenarios.
- The dataset facilitates the study of efficiency, effectiveness, class imbalance, concept drift, and interpretability in AML.
Conclusions:
- SynthAML effectively addresses the limitations posed by the lack of public data in AML research.
- The synthetic dataset is a valuable tool for developing and validating new anti-money laundering strategies and technologies.
