EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Counterfactual Explanations for AML Risk Scores Under Causal Constraints

Ethan Morales, Sophia Bennett, Michael Turner

Frontiers in Business and Finance · 2026

DOI: 10.71465/fbf598 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Frontiers in Business and Finance, 2026. Verfasst von Ethan Morales und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This study introduces a counterfactual-explanation framework for anti-money-laundering (AML) risk scoring. The method distinguishes actionable from non-actionable variables within a causal graph and generates minimal valid changes needed to shift a case below a risk threshold. Experiments used 42.7 million transactions and 3.1 million customer profiles from a nationwide financial dataset. The system generated valid counterfactuals for 91.5% of high-risk cases, requiring an average of 2.4 feature changes per case. In a user evaluation with 26 compliance analysts, counterfactual explanations reduced review time by 22.6% and increased decision consistency from 0.62 to 0.75 (Cohen’s kappa). Causal constraints eliminated unrealistic recommendations in 98.6% of generated explanations. The method enhances model transparency while maintaining detection performance.

Zitieren

APA

Morales, E., Bennett, S., & Turner, M. (2026). Counterfactual Explanations for AML Risk Scores Under Causal Constraints. Frontiers in Business and Finance. https://doi.org/10.71465/fbf598

Vancouver

Morales E, Bennett S, Turner M. Counterfactual Explanations for AML Risk Scores Under Causal Constraints. Frontiers in Business and Finance. 2026. doi:10.71465/fbf598

BibTeX
@article{Morales2026,
  author = {Morales, Ethan and Bennett, Sophia and Turner, Michael},
  title = {Counterfactual Explanations for AML Risk Scores Under Causal Constraints},
  journal = {Frontiers in Business and Finance},
  year = {2026},
  doi = {10.71465/fbf598}
}
Kostenlos starten →