Explainable AI Dashboards for Regulatory Governance BI
DOI:
https://doi.org/10.63282/3050-9246.IJETCSIT-V6I2P118Keywords:
Explainable AI, Business Intelligence, Regulatory Compliance, AI Dashboard, Data Governance, Risk Monitoring, AuditabilityAbstract
The Decision Support Systems have to be transparent, auditable and interpretable in the area of re-regulatory compliance, in particular regarding Business Intelligence (BI) uses, for industries such as financial services, healthcare, insurance or public administration. The present research paper proposes an Explainable AI Dashboard Framework for Regulatory Compliance BI system with various dimensions such as Automated Regulatory Compliance Monitoring, Explainable Machine Learning, Regulatory Risk Scoring, Regulatory Audit Trails, Visual Regulatory Decision Making and Data Governance. In essence, the framework is designed to help organisations highlight areas where they don't meet compliance and track any discrepancies, verify AI's results and report back to the regulators or internal auditors with evidence. This includes explainability (e.g. SHAP, LIME), interactive dashboard widgets, 1000s of data and model-outcomes and enterprise data with a model geometry, on which a human can interact in a comprehendible and user-friendly way. The proposed dashboard would enable users to visualize the risk indicators, the information sources, how to read feature level explanations and to get a real-time view of information available regarding compliance. The framework could be leveraged here to address the “black box” issue in BI regarding AI systems to offer improvements in accountability and transparency in the processes. The paper proposes the concept of “Explainable AI dashboards”, which could be extended in future to facilitate the regulatory review and readiness to assist trust building efforts via the stakeholders and a continuous compliance management effort. This is particularly useful for the organisations that have high requirements and sensitivity when making operational decisions and/or are exposed to high risk.
References
[1] A. Adadi and M. Berrada, “Peeking inside the black-box: A survey on explainable artificial intelligence (XAI),” IEEE Access, vol. 6, pp. 52138–52160, 2018.
[2] I. Ahmed, G. Jeon, and F. Piccialli, “From artificial intelligence to explainable artificial intelligence in Industry 4.0: A survey on what, how, and where,” IEEE Transactions on Industrial Informatics, vol. 18, no. 8, pp. 5031–5042, 2022.
[3] B. Bach, E. Freeman, A. Abdul-Rahman, C. Turkay, S. Khan, Y. Fan, and M. Chen, “Dashboard design patterns,” IEEE Transactions on Visualization and Computer Graphics, vol. 29, no. 1, pp. 342–352, 2022.
[4] P. B. de Laat, “Companies committed to responsible AI: From principles towards implementation and regulation?” Philosophy & Technology, vol. 34, no. 4, pp. 1135–1193, 2021.
[5] J. Gerlach, P. Hoppe, S. Jagels, L. Licker, and M. H. Breitner, “Decision support for efficient XAI services: A morphological analysis, business model archetypes, and a decision tree,” Electronic Markets, vol. 32, no. 4, pp. 2139–2158, 2022.
[6] A. Holzinger, B. Malle, A. Saranti, and B. Pfeifer, “Towards multi-modal causability with graph neural networks enabling information fusion for explainable AI,” Information Fusion, vol. 71, pp. 28–37, 2021.
[7] P. Linardatos, V. Papastefanopoulos, and S. Kotsiantis, “Explainable AI: A review of machine learning interpretability methods,” Entropy, vol. 23, no. 1, Art. no. 18, 2020.
[8] D. Minh, H. X. Wang, Y. F. Li, and T. N. Nguyen, “Explainable artificial intelligence: A comprehensive review,” Artificial Intelligence Review, vol. 55, no. 5, pp. 3503–3568, 2022.
[9] E. Owens, B. Sheehan, M. Mullins, M. Cunneen, J. Ressel, and G. Castignani, “Explainable artificial intelligence (XAI) in insurance,” Risks, vol. 10, no. 12, Art. no. 230, 2022.
[10] G. Srivastava, M. S, R. Venkataraman, K. V, and P. N, “A review of the state of the art in business intelligence software,” Enterprise Information Systems, vol. 16, no. 1, pp. 1–28, 2022.
