PREDICTIVE MODELING OF CREDIT DEFAULT RISK USING ENSEMBLE LEARNING

This paper presented an ensemble learning- based framework for predictive modelling of credit default risk. The proposed system achieves improved predictive accuracy and robustness compared to traditional models.   Future work includes integrating deep neural ensemble architectures, incorporating real-time transaction data for dynamic credit scoring, implementing federated learning for privacy- preserving financial modelling, and enhancing explainability for regulatory transparency. This proctoring is 90% accuracy.  

[1]  Y. Chen et al., “Gradient Boosting Models for Credit Default Risk Prediction,” IEEE Access, 2023.   [2]     M. Kumar et al., “Hybrid Ensemble Learning Framework for Financial Risk Assessment,” IEEE Transactions on Artificial Intelligence, 2024. [3]      L. Zhang et al., “Imbalance Aware Ensemble Methods for Credit Scoring,” IEEE Transactions on Neural Networks and Learning Systems, 2024.   [4]  R. Patel et al., “Explainable AI in Financial Credit Risk Modelling,” IEEE Access, 2025.   [5]       H. Lee et al., “Deep Ensemble Architectures for Financial Risk Prediction,” IEEE Transactions on Artificial Intelligence, 2025. [6]     A. Singh et al., “Scalable Big Data Analytics for Real-Time Credit Scoring,” IEEE Internet of Things Journal, 2026.   [7]    T. Nguyen et al., “Feature Engineering Techniques for Credit Risk Prediction,” IEEE Access, 2023.   [8]   P. Wang et al., “Cost Sensitive Learning for Financial Classification Problems,” IEEE Transactions on Artificial Intelligence, 2024.   [9]      K. Roy et al., “Advanced Boosting Techniques for Banking Analytics,” IEEE Access, 2025. [10]     J. Brown et al., “AI-Based Decision Support Systems in Finance,” IEEE Access, 2026.