Harnessing machine learning for money laundering detection: a criminological theory-centric approach
| Date | 30 December 2024 |
| Pages | 184-201 |
| DOI | https://doi.org/10.1108/JMLC-04-2024-0083 |
| Published date | 30 December 2024 |
| Author | Syahril Ramadhan |
Harnessing machine learning for
money laundering detection: a
criminological theory-centric approach
Syahril Ramadhan
Department of Accounting, Universitas Jakarta International, Jakata, Indonesia and
PostGraduated School of Economics, Pancasila University, Jakarta, Indonesia
Abstract
Purpose –The purpose of this study is to develop and evaluatethe effectiveness of the criminology-centric
machine learning(CCTML) framework in detecting money laundering activitiesby integrating criminological
theorieswith machine learning techniques.
Design/methodology/approach –This study uses a mixed-methods approach, this research synthesizes
qualitative insights from expert interviews and literature reviews with quantitative analysis using mac hine
learning models. Criminology-centric features are engineered based on established theories to capture behaviors
indicative of money laundering. Various machine learning algorithms, including Voting Ensemble, XGBoost,
Random Forest and LightGBM, are evaluatedfor their effectiveness in detecting financial crimes.
Findings –The findings of the study demonstrate that the CCTML approach consistently outperforms
common machine learningmodels in detecting money laundering activities across various evaluation metrics,
including area under the curve, log loss, Matthews correlation coefficient, precision, recall and balanced
accuracy.The integration of criminological insightsinto machine learning models significantly enhancestheir
predictiveaccuracy and reliability.
Originality/value –This research synthesizes diverse criminological insights int o a cohesive framework known
as CCTML. This approach goes beyond common feature engineering by incorporating complex behavi oral patterns
and social dynamics, thereby enhancing the accuracy and transparency of money laundering de tection systems. By
leveraging state-of-the-art machine learning algorithms and explainable artificial intelligence (AI) techniques,
CCTML not only improves predictive capabilities but also ensures that model decisions are interpretable and fair.
Explainable AI helps CCTML reveal why certain transactions are flagged, aiding investigators in ident ifying key
suspects. Furthermore, this study contributes a comprehensive anti-money laundering framew ork that integrates
ethical considerations, promoting a more robust and just approach to combating financial crimes.
Keywords Criminology-centric machine learning (CCTML), Money laundering detection,
Machine learning, Criminological theories, Predictive modeling, Financial crime
Paper type Research paper
1. Introduction
In the age of intricate financial crimes that corrode the very fabric of global economies, the need
for innovative approaches to combat money laundering stands as a pivotal challenge. Levi (2020)
highlights the dual challenge of addressing money laundering as a distinct problem and the
underlying criminal activities that generate illicit funds, assessing both the scale and harmfulness
of these issues, and identifying any emerging challenges in the current anti-money laundering
The author would like to express the deepest gratitude to all the officials and staff at PPATK
(Indonesian Financial Transaction Reports and Analysis Center) for their invaluable support and
assistance in completing this article.
JMLC
28,1
184
Journalof Money Laundering
Control
Vol.28 No. 1, 2025
pp. 184-201
© Emerald Publishing Limited
1368-5201
DOI 10.1108/JMLC-04-2024-0083
The current issue and full text archive of this journal is available on Emerald Insight at:
https://www.emerald.com/insight/1368-5201.htm
(AML) framework. Estimates indicate that illicit funds, stealthily funneled into legitimate
channels annually, amount to billions of dollars, constituting a significant portion of the world
economy (UNODC, 2009). Despite the widespread recognition of money laundering’s
detrimental impacts, there remains a critical gap in understanding the specific tactics and methods
used by criminals, particularly in emerging financial sectors (Tiwari et al., 2023).
This gap is further exacerbated by the convergence of financial sophistication and evolving
criminal methodologies, leading to a paradigm shift in the effectiveness of traditional AML
measures (Gilmour, 2023). Law enforcement agencies have intensified their focus on thwarting
money laundering activities. However, the surge in intricate schemes has laid bare the
inadequacies of conventional prevention strategies, exposing vulnerabilities within the system
(Yeoh,2019;Zdanowicz, 2009). This dynamic landscape calls for innovative solutions capable
of navigating the intricate web of money laundering’s complexities, integrating criminological
theories with advanced machine learning (ML) techniques to enhance detection efficacy.
Recent advancements in technology have introduced new dimensions to financial crimes, such
as money laundering using virtual currency, which complicates detection and regulation efforts
(Webe r et al., 2019). The ongoing evolution of financial crime tactics includes sophisticated
schemes like the “pitch-butchering”typology, where organized crime groups exploit human
trafficking victims to carry out scams (Interpol, 2024). This is just one example of how financial
crime methodologies continue to develop and adapt. Leveraging the evidence-based policing
matrix (Lum et al., 2010) and social network analysis (Fronzetti Colladon and Remondi, 2017)
provides new methodologies for studying crime patterns and identifying criminal networks,
enhancing our ability to detect and prevent these ever-evolving financial crime.
Existing research on money laundering reveals several limitations hindering a holistic
understanding of this multifaceted criminal phenomenon(Kruisbergen et al., 2019). Studies
often delve into specific facets of money laundering, offering fragmented insights into
offender behavior, typologies and methods −akin to examining individual puzzle pieces
without the complete picture. This narrow perspective limits our ability to comprehend the
entirety of money laundering’s intricacies. Moreover, rapid technological advancements
have outpaced conventional researchmethodologies, empowering money launderers to craft
sophisticated, detection-evading schemes.
In the pursuit of more effective detection methods, a multitude of approaches have been
proposed. The Financial Action Task Force (FATF) has underscored the pivotal role of
technology in profiling and detecting moneylaundering activities, heightening the necessity
for sophisticated solutions (FATF, 2021). The ascent of big data analytics, natural language
processing and distributed ledger technology marks the evolving landscape of AML efforts
(Grint et al., 2017;Kaminskiand Schonert, 2018). Amidst these advancements, ML emerges
as a standout, possessing the prowess to handle vast volumes of data, structured and
unstructured alike, unraveling the intricate patterns of illicit financial behaviors (Banwo,
2018;Fernandez, 2019). Recent studies (Jullum et al., 2020) showcase promising results in
ML applications for detectingmoney laundering, providing the impetus to modernizecurrent
AML efforts (Chen etal.,2018;Weber et al., 2018).
However, industry caution persists, recognizing that current models often fall short in
capturing the nuanced cognitive processes, situational contexts and behavioral intricacies
inherent in money launderingactivities (Lokanan, 2024). In response to these challenges, the
development of a criminological theory-centric typology of money laundering (CCTML)
emerges as a pressing necessity. This integration aims to revolutionize the landscape of
financial crime detectionand enforcement by leveraging insights from criminology.
Moreover, the integration of explainable AI (XAI) and Algorithmic Justice within the
CCTML framework is crucial (Samek et al., 2019). These conceptsensure that ML models
Journal of Money
Laundering
Control
185
Get this document and AI-powered insights with a free trial of vLex and Vincent AI
Get Started for FreeStart Your Free Trial of vLex and Vincent AI, Your Precision-Engineered Legal Assistant
-
Access comprehensive legal content with no limitations across vLex's unparalleled global legal database
-
Build stronger arguments with verified citations and CERT citator that tracks case history and precedential strength
-
Transform your legal research from hours to minutes with Vincent AI's intelligent search and analysis capabilities
-
Elevate your practice by focusing your expertise where it matters most while Vincent handles the heavy lifting
Start Your Free Trial of vLex and Vincent AI, Your Precision-Engineered Legal Assistant
-
Access comprehensive legal content with no limitations across vLex's unparalleled global legal database
-
Build stronger arguments with verified citations and CERT citator that tracks case history and precedential strength
-
Transform your legal research from hours to minutes with Vincent AI's intelligent search and analysis capabilities
-
Elevate your practice by focusing your expertise where it matters most while Vincent handles the heavy lifting
Start Your Free Trial of vLex and Vincent AI, Your Precision-Engineered Legal Assistant
-
Access comprehensive legal content with no limitations across vLex's unparalleled global legal database
-
Build stronger arguments with verified citations and CERT citator that tracks case history and precedential strength
-
Transform your legal research from hours to minutes with Vincent AI's intelligent search and analysis capabilities
-
Elevate your practice by focusing your expertise where it matters most while Vincent handles the heavy lifting
Start Your Free Trial of vLex and Vincent AI, Your Precision-Engineered Legal Assistant
-
Access comprehensive legal content with no limitations across vLex's unparalleled global legal database
-
Build stronger arguments with verified citations and CERT citator that tracks case history and precedential strength
-
Transform your legal research from hours to minutes with Vincent AI's intelligent search and analysis capabilities
-
Elevate your practice by focusing your expertise where it matters most while Vincent handles the heavy lifting
Start Your Free Trial of vLex and Vincent AI, Your Precision-Engineered Legal Assistant
-
Access comprehensive legal content with no limitations across vLex's unparalleled global legal database
-
Build stronger arguments with verified citations and CERT citator that tracks case history and precedential strength
-
Transform your legal research from hours to minutes with Vincent AI's intelligent search and analysis capabilities
-
Elevate your practice by focusing your expertise where it matters most while Vincent handles the heavy lifting