The rapid evolution of financial technology (FinTech) has significantly transformed the landscape of microcredit risk assessment, particularly in developing economies where financial inclusion remains a persistent challenge. Traditional credit scoring systems rely heavily on formal financial histories, thereby excluding a large segment of the population that lacks access to formal banking services. While machine learning (ML) models have improved predictive accuracy by leveraging alternative data sources such as mobile usage, transaction behavior, and social indicators, their inherent lack of transparency poses serious concerns related to fairness, accountability, and regulatory compliance. This study investigates the role of Explainable Artificial Intelligence (XAI) in enhancing transparency in microcredit risk assessment without compromising predictive performance. By integrating advanced ML algorithms such as Random Forest and Gradient Boosting with interpretability tools like SHAP (SHapley Additive exPlanations), the research evaluates both model accuracy and explainability. The study adopts a mixed-method approach, utilizing primary data collected through surveys and interviews, along with secondary datasets from financial institutions and global databases. The findings reveal that XAI-enabled models can maintain high predictive accuracy while significantly improving interpretability and stakeholder trust. The research contributes to the existing literature by proposing a hybrid framework that aligns technological innovation with ethical and regulatory requirements, thereby fostering inclusive financial systems.
Devi et al. (Wed,) studied this question.