Utilizing Explainable AI in Financial Risk Assessment: Enhancing User Empowerment through Interpretable Credit Scoring Models

This research presents a framework that integrates machine learning methods with Explainable Artificial Intelligence (XAI) to enhance predictive accuracy and interpretability in risk assessment. Models ranging from Logistic Regression and Decision Trees to ensemble methods like Random Forest, CatBoost, and XGBoost are developed to address credit risk challenges and enhance decision-making transparency. Although the approach is adaptable for financial risk modeling, the framework is tested using the abalone dataset from the UCI Machine Learning Repository, which includes physical measurements and categorical attributes for performance evaluation. Rigorous preprocessing with feature engineering and standardization ensures data integrity. Model performance, measured by Mean Squared Error (MSE), shows that Random Forest and CatBoost yield superior results. Additionally, Principal Component Analysis (PCA) and feature importance analysis improve interpretability by revealing key predictive factors. Overall, the results underscore the viability of integrating XAI with machine learning for risk assessment.