Deep Learning and Explainable Artificial Intelligence Techniques Applied for Detecting Money Laundering–A Critical Review
Open accessThe DOI matches its Crossref record; 92 of the 92 references it cites were found in a registry or an index.
Abstract
Money laundering has been a global issue for decades, which is one of the major threat for economy and society. Government, regulatory and financial institutions are combating it together in their respective capacity, however still billions of dollars in fines by authorities make the headlines in the news. High-speed internet services have enabled financial institutions to deliver better customer experience through multi-channel engagements, which has led to exponential growth in transactions and new avenues for laundering the money for fraudsters. Literature shows the usage of statistical methods, data mining and Machine Learning (ML) techniques for money laundering detection, but limited research on Deep Learning (DL) techniques, primarily due to lack of model interpretability and explainability of the decisions made. Several studies are conducted on application of ML for Anti-Money Laundering (AML), and Explainable Artificial Intelligence (XAI) techniques in general, but lacks the study on usage of DL techniques together with XAI. This paper aims to review the current state-of-the-art literature on DL together with XAI for identifying suspicious money laundering transactions and identify future research areas. Key findings of the review are, researchers have preferred variants of Convolutional Neural Networks, and AutoEncoder; graph deep learning together with natural language processing is emerging as an important technology for AML; XAI use is not seen in AML domain; 51% ML methods used in AML are non-interpretable, 58% studies used sample of old real data; key challenges for researchers are access to recent real transaction data and scarcity of labelled training data; and data being highly imbalanced. Future research directions are, application of XAI techniques to bring-out explainability, graph deep learning using natural language processing (NLP), unsupervised and reinforcement learning to handle lack of labelled data; and joint research programs between research community and industry to benefit from domain knowledge and controlled access to data.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Figures
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The cited source says what the claim attributes to it.
In the citing paper
“Explainable artificial intelligence [12] offers both knowledgedriven and data-driven interpretability approaches [96].”
From A 2030 Roadmap for Software Engineering (2025) · cites this paper as [96]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Going deeper with convolutions
Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V., & Rabinovich, A. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298594 · Cited by 48,562
Semi-supervised classification with graph convolutional networks
Kipf, T. N., & Welling, M. (2016) · arXiv · doi:10.48550/arxiv.1609.02907 · Cited by 24,853 · cited 1×
“First using the graph model GCN developed by [37], which took 611 seconds to converge in 32”
MethodA Unified Approach to Interpreting Model Predictions
Lundberg, S., & Lee, S.-I. (2017) · arXiv · doi:10.48550/arxiv.1705.07874 · Cited by 23,707 · cited 1×
“Local Interpretable Model-Agnostic Explanations (LIME) [99] and SHapley Additive exPlana-tions (SHAP) [100] are the highly referred and experimented methods for producing explanations on black box models that use any type of data, followed by Partial Dependence Plots (PDPs).”
Method"Why Should I Trust You?": Explaining the Predictions of Any Classifier
Ribeiro, M. T., Singh, S., & Guestrin, C. (2016) · Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining · doi:10.1145/2939672.2939778 · Cited by 17,555
Gauriat, C.-M., Pencolé, Y., Ribot, P., & Brouillet, G. (2024) · HAL (Le Centre pour la Communication Scientifique Directe) · doi:10.4230/oasics.dx.2024.27 · Cited by 12,586
Arrieta, A. B., Díaz-Rodríguez, N., Del Ser, J., Bennetot, A., Tabik, S., Barbado, A., García, S., Gil-López, S., Molina, D., Benjamins, R., et al. (2019) · arXiv · doi:10.48550/arxiv.1910.10045 · Cited by 11,321 · cited 2×
“by these models to conclude a transaction is suspicious; and a mandatory ask by regulators to provide rationale of transaction being suspicious, is posing an impediment for adoption of AI/ML technology in finance crime units in financial institutes [16].”
MethodRudin, C. (2019) · Nature Machine Intelligence · doi:10.1038/s42256-019-0048-x · Cited by 10,561 · cited 1×
“While there is mammoth progress on XAI method, [102] argues that the development in XAI is a patchwork to existing ML models and major effort of research community is going in the wrong direction.”
MethodBarredo Arrieta, A., Díaz-Rodríguez, N., Del Ser, J., Bennetot, A., Tabik, S., Barbado, A., Garcia, S., Gil-Lopez, S., Molina, D., Benjamins, R., et al. (2020) · Information Fusion · doi:10.1016/j.inffus.2019.12.012 · Cited by 10,423
Peeking Inside the Black-Box: A Survey on Explainable Artificial Intelligence (XAI)
Adadi, A., & Berrada, M. (2018) · IEEE Access · doi:10.1109/access.2018.2870052 · Cited by 6,294 · cited 1×
“…wants the explanation to justify the decisions, security application domain wants the explanation to control the underlying data and IP that may be uncovered, some domains would need the explanation to improve the decisions, other domains would need the explanation to discover the knowledge [90].”
BackgroundExplainable AI: a review of machine learning interpretability methods
Linardatos, P., Papastefanopoulos, V., & Kotsiantis, S. (2020) · Entropy · doi:10.3390/e23010018 · Cited by 2,922 · cited 1×
“[97] presented a comprehensive review of machine learning interpretabilitymethods using 4 categories– (i) Methods for explaining black box models (17 methods focused on DL method interpretations, 16 methods that can”
Method
Research works printed but not linked: 17
- J. Han, ''NextGen AML: Distributed deep learning based language tech- nologies to augment anti money laundering investigation,'' in Proc. Assoc. Comput. Linguistics (ACL), 2015, pp. 37-42.
- B. Kitchenham and S. Charters, ''Guidelines for performing systematic literature reviews in software engineering,'' in Guidelines for Performing Systematic Literature Reviews in Software Engineering.
- B. Shaju and N. Valliammal, ''Measures for financial fraud detection using data analytics and machine learning,'' Int. J. Adv. Sci. Technol., vol. 28, no. 17, pp. 270-280, 2019.
- A. Semenov, ''Survey of common design approaches in AML software development,'' in Proc. CEUR-WS, 2017, pp. 1-9.
- V. Jayasree and R. V. S. Balan, ''Anti money laundering in financial institutions using affiliation mapping calculation and sequential mining,'' J. Eng. Appl. Sci., vol. 11, no. 1, pp. 51-56, 2016.
- Bank Secrecy Act (BSA) Statute, F.C.E. Netw., Vienna, Austria, Feb. 2021.
- CBS News. (2012). HSBC to Pay $1.9B to Settle Money-Laundering Case. www.cbc.ca [Online]. Available: https://www.cbc.ca/news/business/hsbc- to-pay-1-9b-to-settle-money-laundering-case-1.1226871
- I. Alarab, S. Prakoonwit, and M. Nacer, ''Competence of graph convolu- tional networks for anti-money laundering in bitcoin blockchain,'' in Proc. 5th Int. Conf. Mach. Learn. Technol. (ICMLT), 2020, pp. 23-27. · DOI
- E. L. Paula, ''Deep learning anomaly detection as suppor fraud investiga- tion in Brazilian exports and anti-money laundering,'' in Proc. 15th IEEE Int. Conf. Mach. Learn. Appl. (ICMLA), 2017, pp. 954-960.
- C. H. Zhang and X. H. Zhao, ''Research on money laundering recognition based on decision thee algorithm,'' Wuhan Ligong Daxue Xuebao/J. Wuhan Univ. Technol., vol. 30, no. 2, pp. 154-156, 2008.
- 2021, Money-laundering and globalization
- F. A. T. Force. (2020). International Standards On Combating Money Laundering and The Financing Of Terrorism & Proliferation-FATF Recommendations, F.A.T. Force, Editor. 2020, Financial Action Task Force. [Online]. Available: http://www.fatf-gafi.org/publications/fatfrecommendations/documents/ the40recommendationspublishedoctober2004.html
- F. S. Board, Artificial Intelligence and Machine Learning in Financial Services 2017. Basel, Switzerland: Financial Stability Board.
- B. Monroe. (Feb. 2021). Fincrime Briefing: AML Fines in 2019 Breach $8 Billion, Treasury Official Pleads Guilty to Leaking, 2020 Crypto Compliance Outlook. [Online]. Available: https://www.acfcs.org/ fincrime-briefing-aml-fines-in-2019-breach-8-billion-treasury- official-pleads-guilty-to-leaking-2020-crypto-compliance- outlook-and-more/#:~:text=Key%20observations%3A,25%20 penalties%20totaling%20%242.29bn
- Media Company. (2019). Risk Transforming Approaches to AML and Financial Crime. Accessed: Feb. 26, 2021. [Online]. Available: https://www.mckinsey.com/~/media/McKinsey/Business%20Functions/ Risk/Our%20Insights/Transforming%20approaches%20to%20AML%20 and%20financial%20crime/Transforming-approaches-to-AML-and- financial%20crime-vF.pdf
- Institute of International Finance. (2018). Machine Learning in Anti- Money Laundering-Summary Report. [Online]. Available: www.iif.com
- PricewaterhouseCoopers. (Feb. 2021). Explainable AI Driving Busi- ness Value Through Greater Understanding. 2017. [Online]. Avail- able: https://www.pwc.co.uk/services/risk-assurance/insights/explainable- ai.html
Cited by
Items 201–209 of 209
Strengthening Transaction Security in Financial Systems Using Edge Artificial Intelligence
Mendiratta, A., Ali, S. B. S., Saxena, R., Verma, A., Shankar, S., & Kumar, R. P. (2026) · OPJU International Technology Conference on Smart Computing for Innovation and Advancement in Industry (OTCON) · doi:10.1109/otcon69332.2026.11629990
“It is proved by the research that intelligent models are more efficient than the system based on fixed rules in detecting complex patterns of fraud and new attacks that have not been identified before [13].”
Shafiee, S., Estilaiee, M., & Dastyar, S. A. R. (2025) · Iranian Joint Congress on Fuzzy and Intelligent Systems · doi:10.1109/cfis68949.2025.11652311
Bhattarai, S. S., Bhandari, S., Ji, B., & Kim, H. (2026) · ASCE Construction Research Congress (CRC) · doi:10.1061/9780784486962.077
Dual Use of Emerging Technologies and Decentralised Financial Assets
Adamyk, B., & Benson, V. (2026) · Cambridge University Press eBooks · doi:10.1017/9781009623605.008
Arpitha, Shetti, P. P., Patil, P. S., & Mathews, R. A. (2026) · Lecture notes in networks and systems · doi:10.1007/978-3-032-25087-2_12
A Review of Deep Learning Models and Explainable Artificial Intelligence for Pathology Recognition
Toala-Paz, A., Becerra-Andrade, A., Cevallos-Levicek, D., Peláez, E., Domínguez, F., & Loayza, F. R. (2026) · International Conference on eDemocracy & eGovernment · doi:10.1109/icedeg70169.2026.11695396
“They focus on the use of learning models that have a simple structure and are easily un-derstandable, such as Decision Trees (DT), Linear Regression (LR), or rule-based models [25].”
Jiao, P. (2026) · Discover Computing · doi:10.1007/s10791-026-10625-9
“However, given its reliance on simulated fraud data and unsupervised criteria, the system is a limited representation of all real-world money laundering patterns and no insight into how the system determined its decisions (as is typical with autoencoder/GAN black box models) [14].”
Design of a Secure Financial Analytics Framework for Risk Assessment and Regulatory Adherence
Febiula, I. M. C., Lakshmi, A. U., Vijayakumar, S., & Ramesh, R. (2026) · IEEE World Conference on Applied Intelligence and Computing (AIC) · doi:10.1109/aic70896.2026.11716913
“The data processed is converted into meaningful insights and reports as part of analytics and reporting [7].”
Fu, H., Inha, Z., & Zhang, Y. (2026) · Communications in computer and information science · doi:10.1007/978-981-92-7298-3_14
Metadata
- Authors
- Kute, D. V., Pradhan, B., Shukla, N., & Alamri, A.
- Affiliations
- University of Technology Sydney (AU)
- King Saud University (SA)
- Venue
- IEEE Access
- Published
- 2021
- Volume
- 9
- Pages
- 82300-82317
- ISSN
- 2169-3536 (ISSN-L)
- OpenAlex
- W3171973514
- Cited by
- 209
- References
- 89
- Citations by year
- 2026: 45 · 2025: 65 · 2024: 46 · 2023: 34 · 2022: 16 · 2021: 3
- FWCI
- 8.71
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Locations
- IEEE Access (publishedVersion, cc-by)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Griffith Research Online (submittedVersion, cc-by)
- Greater South Information System
- Funders
- Centre for Advanced Modelling and Geospatial Information Systems (CAMGIS), Faculty of Engineering and IT, University of Technology Sydney (UTS), Sydney, NSW, Australia
- Researchers Supporting Project, King Saud University, Riyadh, Saudi Arabia (RSP-2020/14)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Imbalanced Data Classification Techniques
- Social Sciences › Sociology and Political Science › Crime, Illicit Activities, and Governance
- Physical Sciences › Computer Science › Artificial Intelligence › Explainable Artificial Intelligence (XAI)
- Keywords
- money laundering detection, anti-money laundering, deep learning, explainable AI, convolutional neural networks, autoencoder, graph-based deep learning, natural language processing, model interpretability, class imbalance, financial transaction data, fraud detection
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.








