Authors: Irakshetti Pranav Amol, Prof. C. A. Kolse, Prof. M. C. Kolse, Prof. S. K. Said
Abstract: Credit risk assessment is an important process in the banking and financial industry, as it helps financial institutions evaluate the possibility of borrowers failing to repay their loans. With the advancement of Artificial Intelligence (AI) and Machine Learning (ML), traditional credit assessment methods are increasingly being supported by intelligent prediction models. However, many advanced AI models operate as black-box systems, making it difficult to understand how they arrive at particular credit decisions. Explainable Artificial Intelligence (XAI) addresses this limitation by providing transparency and interpretability in AI based credit risk assessment. This paper presents a review of existing XAI techniques and their applications in predicting credit risk and supporting financial decision-making. Various machine learning approaches, including Decision Trees, Random Forest, Logistic Regression, and Gradient Boosting, are examined along with explainability techniques such as SHAP (SHapley Additive exPlanations), LIME (Local Interpretable Model-Agnostic Explanations), and feature importance analysis. The review discusses how these methods help identify important factors influencing credit scores, loan approvals, and default predictions. It also highlights the benefits of explainable models, including improved transparency, customer trust, accountability, and regulatory compliance. Furthermore, major challenges such as data privacy, algorithmic bias, model complexity, and the balance between prediction accuracy and interpretability are explored. The paper also discusses future research opportunities in developing fair, reliable, and transparent AI-based financial systems. Overall, this review emphasizes the importance of Explainable AI in making credit risk assessment more understandable, trustworthy, and responsible while supporting better financial decisions.
DOI: https://doi.org/10.5281/zenodo.24214153
