The increasing integration of Artificial Intelligence (AI) in financial services has significantly optimized the loan approval process. However, the lack of transparency and interpretability in AI decisions has raised concerns regarding trust, fairness, and accountability. This paper proposes a novel Explainable Artificial Intelligence (XAI) framework for predicting loan approval status and rejection reasons, thereby enhancing stakeholder trust in AI-driven financial decisions. The model utilizes SHAP (SHapley Additive exPlanations) values to interpret the contribution of each feature in classification tasks, offering granular insights into why a particular loan application is approved or denied. An experimental analysis of a real-world loan application dataset reveals that the proposed model achieves a high prediction accuracy of 90% in identifying rejection reasons, while maintaining explainability through both visual and numerical interpretations. The results demonstrate the effectiveness of XAI in making complex AI models interpretable and regulatory-compliant. This work contributes to building transparent, ethical, and reliable financial systems by integrating AI with human-understandable justifications for decisions.
Sudiptha et al. (Tue,) studied this question.