Shapley Value-Guided Adaptive Ensemble Learning for Explainable Financial Fraud Detection with U.S. Regulatory Compliance Validation

2026-04-14

AI Summary

This study addresses the challenge that existing financial fraud detection models fail to meet stringent U.S. regulatory requirements—such as OCC Bulletin 2011-12 and SR 11-7—regarding model interpretability and auditability. To bridge this gap, the authors propose a SHAP value-guided adaptive ensemble (SGAE) method that dynamically optimizes transaction-level model weights, achieving high detection performance while delivering transparent, regulation-compliant explanations. The work introduces a novel SHAP-driven ensemble mechanism and systematically evaluates the fidelity and stability of explanations across diverse architectures, including XGBoost, LSTM, Transformer, and GNN-GraphSAGE. Evaluated on the IEEE-CIS dataset, SGAE attains a test AUC-ROC of 0.8837, with GNN-GraphSAGE emerging as the top performer (AUC-ROC: 0.9248; F1: 0.6013). All findings are explicitly mapped to compliance frameworks under OCC, SR 11-7, and BSA-AML regulations.

Technology Category

• Machine Learning: Transparent, Interpretable, Explainable ML
• Humans and AI: Explainable AI (XAI) for Human Understanding
• Philosophy and Ethics of AI: Accountability, Interpretability & Explainability

Application Category

• User Modeling, Personalization and Recommendation: Accountability, Transparency, and Ethics for personalization
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

Financial crime costs U.S. institutions over $32 billion each year. Although AI tools for fraud detection have become more advanced, their use in real-world systems still faces a major obstacle: many of these models operate as black boxes that cannot provide the transparent, auditable explanations required by regulations such as OCC Bulletin 2011-12 and Federal Reserve SR 11-7. This study makes three main contributions. First, it offers a thorough evaluation of explanation quality across faithfulness (sufficiency and comprehensiveness at k=5, 10, and 15) and stability (Kendall's W across 30 bootstrap samples). XGBoost paired with TreeExplainer achieves near-perfect stability (W=0.9912), while LSTM with DeepExplainer shows weak results (W=0.4962). Second, the paper introduces the SHAP-Guided Adaptive Ensemble (SGAE), which dynamically adjusts per-transaction ensemble weights based on SHAP attribution agreement, achieving the highest AUC-ROC among all tested models (0.8837 held-out; 0.9245 cross-validation). Third, a complete three-architecture evaluation of LSTM, Transformer, and GNN-GraphSAGE on the full 590,540-transaction IEEE-CIS dataset is provided, with GNN-GraphSAGE achieving AUC-ROC 0.9248 and F1=0.6013. All results are mapped directly to OCC, SR 11-7, and BSA-AML regulatory compliance requirements.

Problem

• Explainable AI
• Financial Fraud Detection
• Regulatory Compliance
• Black-box Models
• SHAP

Innovation

• SHAP-Guided Adaptive Ensemble
• Explainable AI
• Financial Fraud Detection
• Regulatory Compliance
• Model Interpretability