Enhancing AML compliance: a machine learning approach to suspicious activity detection through routine activity theory
Bibliographic record
Abstract
Purpose This study explores the application of machine learning (ML) algorithms to enhance the detection and reporting of Suspicious Activity Reports (SARs) in California’s financial sector. This research aims to improve anti-money laundering (AML) compliance by evaluating the effectiveness of advanced ML techniques, specifically CatBoost and Decision Tree algorithms, in identifying suspicious financial transactions. Design/methodology/approach This research uses a comprehensive methodological framework involving the analysis of 45,000 SAR filings from financial institutions and regulatory agencies in California, dating back to 2018. Various ML algorithms, including linear regression, random forest, decision tree and CatBoost, are used to analyze SAR filing patterns and predict suspicious transactions. Findings The findings reveal that CatBoost outperforms other models, offering a better fit to the data and higher predictive accuracy with a low RMSE and high cross-validation scores. The Decision Tree algorithm also demonstrates significant promise but is slightly less effective than CatBoost. This study confirms that ML algorithms, particularly CatBoost, significantly improve the detection and reporting of suspicious financial activities, thereby enhancing AML compliance. Originality/value This research contributes to the literature by integrating advanced ML techniques into AML compliance, moving beyond traditional statistical approaches. The findings provide practical implications for financial institutions, highlighting the potential of ML algorithms to enhance the effectiveness of SAR filings and bolster regulatory efforts in mitigating financial crime. This study underscores the value of ML in developing targeted policies to curb illicit financial activities and advance AML analytical capabilities.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.006 | 0.028 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.004 | 0.002 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".