Detecting money laundering transactions with machine learning
Pages | 173-186 |
DOI | https://doi.org/10.1108/JMLC-07-2019-0055 |
Published date | 21 January 2020 |
Date | 21 January 2020 |
Author | Martin Jullum,Anders Løland,Ragnar Bang Huseby,Geir Ånonsen,Johannes Lorentzen |
Subject Matter | Financial risk/company failure,Financial compliance/regulation,Accounting & Finance |
Detecting money laundering
transactions with
machine learning
Martin Jullum,Anders Løland and Ragnar Bang Huseby
Norwegian Computing Center, Oslo, Norway, and
Geir Ånonsen and Johannes Lorentzen
DNB, Oslo, Norway
Abstract
Purpose –The purpose of this paper is to develop, describe and validate a machine learning model for
prioritising which financial transactions should be manually investigated for potential money laundering.
The model is appliedto a large data set from Norway’s largest bank, DNB.
Design/methodology/approach –A supervisedmachine learning model is trained by using three types
of historic data: “normal”legaltransactions; those flagged as suspicious by the bank’s internal alert system;
and potential money laundering cases reported to the authorities. The model is trained to predict the
probability that a new transaction should be reported, using information such as background information
about the sender/receiver,their earlier behaviour and their transaction history.
Findings –The paper demonstrates that the common approach of not using non-reported alerts (i.e.
transactions that are investigated but not reported) in the training of the model can lead to sub-optimal
results. The same applies to the use of normal (un-investigated) transactions. Our developed method
outperformsthe bank’s current approach in terms of a fair measure of performance.
Originality/value –This research study is one of very few published anti-money laundering (AML)
models for suspicious transactions that have been applied to a realistically sized data set. The paper also
presents a new performance measure specifically tailored to compare the proposed method to the bank’s
existingAML system.
Keywords Machine learning, XGBoost, Supervised learning, Suspicious transaction
Paper type Research paper
1. Introduction
The true extent of money laundering transactions is unknown and uncertain, potentially
because financial firms lack incentiveand tools to estimate the extent of money laundering
in their accounts (Reuter and Truman, 2004). In an old report to US Congress (1995), it was
estimated that about 0.05-0.1per cent of the transactions through the Society for Worldwide
Interbank Financial Telecommunications system involved money laundering. A meta-
analysis by United Nations Office on Drugs and Crime (2011) estimates that the total
amount of money laundered through the financialsystem is equivalent to about 2.7 per cent
© Martin Jullum, Anders Løland, Ragnar Bang Huseby, Geir Ånonsen and Johannes Lorentzen.
Published by Emerald Publishing Limited. This article is published under the Creative Commons
Attribution (CC BY 4.0) licence. Anyone may reproduce, distribute, translate and create derivative
works of this article (for both commercial and non-commercial purposes), subject to full attribution to
the original publication and authors. The full terms of this licence may be seen at http://
creativecommons.org/licences/by/4.0/legalcode
Money
laundering
transactions
173
Journalof Money Laundering
Control
Vol.23 No. 1, 2020
pp. 173-186
EmeraldPublishing Limited
1368-5201
DOI 10.1108/JMLC-07-2019-0055
The current issue and full text archive of this journal is available on Emerald Insight at:
https://www.emerald.com/insight/1368-5201.htm
To continue reading
Request your trial