Anti-money laundering systems: a systematic literature review

Publication Date25 May 2020
AuthorAlhanouf Abdulrahman Saleh Alsuwailem,Abdul Khader Jilani Saudagar
SubjectAccounting & Finance,Financial risk/company failure,Financial compliance/regulation,Financial crime
Anti-money laundering systems: a
systematic literature review
Alhanouf Abdulrahman Saleh Alsuwailem
Department of Information Systems, Imam Mohammad Ibn
Saud Islamic University, Riyadh, Saudi Arabia, and
Abdul Khader Jilani Saudagar
Department of Information Systems, College of Computer and Information
Sciences, Imam Mohammad Ibn Saud Islamic University, Riyadh, Saudi Arabia
Purpose This paper aims to understand and document the stateof the art in the anti-money laundering
(AML) systemsliterature.
Design/methodology/approach A systematic literature review (SLR) is performed using the Saudi
Digital Library.The outputs published as conference proceedings,workshop proceedings, journalarticles and
books were all considered. The nal sample size after omitting out-of-scope selections was 27 documents,
which mainlyspan from 2015 to 2020.
Findings The sample is discussed based on a categorization, which demarcates solutions, machine
learning,data sources, evaluation methods, implementationtools, sampling techniques and regions of study.
Originality/value This SLR could serve as a useful basis for researchers and salient decision-makers,
who are seeking to understandthe nature and extent of the currently available research into AML systems.
Keywords Algorithms, Anti-money laundering systems, Detection, Machine learning, Systematic
literature review
Paper type Literature review
1. Introduction
Money laundering is dened as the process of hiding an illicit origin of dirty money and
making it seem legitimate and valid (Le-Khac et al.,2016). ML can also be dened as the
process of cleaning dirtymoney, which means funds collected from criminal or illegal
activities including drug trafcking, illegal gambling and tax evasion (Soltani et al.,2016)
and (Salehi et al., 2017). Another denition of ML is the process of converting
unaccountablemoney into accountable money(Suresh et al.,2016).
ML has a negative impact on the global economy, and it is considered a serious problem
discussed by countries around the world (Syed Mustapha Nazri et al.,2019). Indeed, ML is
the third largest business around the world, accounting for about 2.7% of global gross
domestic product (GDP) after the currency exchange and auto industries (Le-Khac et al.,
2016;Soltani et al., 2016). The International Monetary Fund estimated that ML proceedsare
between 2% and 5% of global GDP (Syed MustaphaNazri et al.,2019). Also, as stated by the
United Nations Ofce on Drug and Crime, the laundered money in one year amounts to
between $500bn and $1tn across the world and of thismoney about $400bn$450bn comes
from drug trafcking (Le-Khac et al., 2016). ML has three main stages: placement, layering
(also known as occultation) and integration. In the rst stage, funds are obtained through
criminal activities and introduced into the nancialsystem. Then, in the layering stage, the
original source of the money is hidden by spreading the money to multiple intermediaries.
Journalof Money Laundering
Vol.23 No. 4, 2020
pp. 833-848
© Emerald Publishing Limited
DOI 10.1108/JMLC-02-2020-0018
The current issue and full text archive of this journal is available on Emerald Insight at:
Finally, in the integration stage, the illegalcash is transferred to the owner (Alexandre and
Balsa, 2015;Salehiet al., 2017;Savage et al.,2017;Soltani et al.,2016;Suresh et al., 2016).
The objectives of the Financial Action Task Force (FATF) group are to provide
standards and enhance legal, regulatory and operational measures to ght ML, terrorist
nancing and any other threats that would affect the probity of the international nancial
system [About - Financial Action Task Force (FATF), 2020]. Generally, most governments
keep looking to improve processesto minimize or prevent illegal activities that affect capital
(Syed Mustapha Nazri et al., 2019). In addition, governments around the world have issued
regulations and recommendations to combat ML. For example, the Financial Transactions
and Reports Analysis Centre in Canada is responsible for establishing regulations and
policies in that country, affecting accountants, banks and real estate organizations. In the
USA, FinCEN provides regulations and recommendations to nancial institutions.
International agencies such as the FATF also set standards and give recommendations for
ghting ML (Soltani et al.,2016).
The detection of ML commenced in 1970, when nancial institutions started notifying
huge transactions to theirgovernments. In the late 1990s, statistical techniques were used to
detect ML patterns, specically Bayesian models and temporal sequence matching, and in
2004, machine learning techniques began being used to the same end. The most popular
algorithms for detecting ML activities are the C4.5 decision tree (DT), support vector
machine (SVM) and radial-based function neural network model (Soltani et al.,2016). A
system that works againstML is called an anti-money laundering (AML) system.
The intention herein is to conduct a systematic literature review (SLR) that mainly
explores existing AML techniques, whichuse machine learning and other methodologies to
detect suspicious transactions. Accordingly, methodological details explaining how articles
were selected for analysis are provided in Section 2. The ndings of the review are then
documented in Section 3 in a structuredway based on classifying the research outputs. The
paper then culminatesby offering conclusions in Section 4.
2. Methodology
2.1 Research scope and focus
The scope of this review is AML techniques and their effectiveness as well as other
approaches proposed by researchers for the detection of suspicious nancial activities. The
main focus is on applications that use machine learning methods, social network analysis,
deep learning and otherAML solutions.
2.2 Research framework
A SLR was performed to survey the current state of the art of AML methods. The SLR is
therefore guided by the following research question: What are the existing methods and
approaches that can be usedto detect ML activities?
The SLR was performed by using the Saudi DigitalLibrary with English databases that
included libraries such asSpringer, IEEE Xplore, ScienceDirect, Emerald Insight andACM.
The documents in the sample mainly span from 2015 to 2020, with two items from 2010 and
one from 2011. The keywords used while searching in the digital library were as follows:
Anti-money laundering, money laundering, detecting money laundering, AML, Anti-money
laundering systems. The nal sample size after omitting out-of-scope selections was 27

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT