This paper introduces an Explainable Artificial Intelligence (XAI) framework designed for anomaly detection in banking transactions, specifically for internal audit purposes. The system utilizes an Isolation Forest model for unsupervised anomaly scoring and a SHAP layer to provide feature-attributed explanations. A Streamlit dashboard makes these outputs accessible to auditors without ML expertise, and evaluations show improved precision and recall compared to baseline methods, with expert feedback indicating enhanced auditor confidence and decision quality. AI
IMPACT Enhances transparency and decision quality in regulated financial environments by making AI outputs interpretable for auditors.
RANK_REASON The cluster contains a research paper detailing a new framework for anomaly detection in banking transactions.
- alphaXiv
- arXiv
- CatalyzeX
- Connected Papers
- DagsHub
- Gotit.pub
- Hugging Face
- Isolation Forest
- Litmaps
- ScienceCast
- scite Smart Citations
- SHAP
- Streamlit
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →