A novel hybrid credit scoring model: integrating RNN and XGBoost for improved creditworthiness assessment

Abstract

Credit scoring is essential for financial management, enabling lenders to assess creditworthiness. Traditional methods often struggle with complex data, leading to suboptimal results. We propose a hybrid credit scoring model combining Recurrent Neural Networks (RNNs) and XGBoost ensemble techniques, enhanced by Randomised Averaging. RNNs process sequential data, capturing long-term dependencies in credit histories, while XGBoost handles structured data patterns. Transfer learning from the Diane dataset adds diverse credit market insights. Randomised Averaging improves predictive performance and reduces overfitting. Experiments on datasets from Australia, Germany, Japan, and Diane show the model's superior accuracy, precision, recall, F1 score, and Brier score, outperforming existing methods and traditional credit scoring. This innovative approach effectively addresses the challenges of low accuracy and high training time in creditworthiness assessment, highlighting its robustness and practical applicability in various financial contexts.

KEYWORDS:

• RNN model
• XGBoost ensemble model
• machine learning
• creditworthiness assessment
• Diane dataset
• K-fold cross-validation

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Annie Chacko

Annie Chacko is pursuing Ph.D in Data Mining in the Department of Computer Science and Engineering at Hindustan Institute of Technology and Science, Chennai. Now, she is working as Associate Professor in the department of Computer Science and Engineering at MBC College of Engineering and Technology, Peermade, Idukki, Kerala. She has received B.Tech in Information Technology in the year 2006 and M.E in Computer Science and Engineering in the year 2012. She has 16 years’ experience in teaching. She has published 1 Scopus indexed journal and attended 2 international conferences and different FDPs, Workshops and Seminars.

David John Aravindhar

David John Aravindhar is currently a Professor in the Department of Computer Science and Engineering, Hindustan Institute of Technology and Science. He received his M.E (Computer Science and Engineering) from Madras University, in 2003 and Ph.D (Computer Science and Engineering) from HITS in 2015. His Ph.D dissertations focussed on Data Mining. He has teaching experience of more than 20 years in the areas of Data Mining, Artificial Intelligence, IoT and Cloud Computing. Currently, he is the Head of Admissions, Hindustan Institute of Technology and Science. He has published over 75 papers in technical journals and conferences.

Arokiasamy Antonidoss

Arokiasamy Antonidoss is working as Professor and Head, in the Department of Computer Science and Business Systems, S.A Engineering College (Autonomous) Chennai Tamilnadu, India. He obtained degrees in B.E and M.E Computer Science and Engineering, Ph.D. He has published papers in peer reviewed journals and two test books in reputed publications. Under his guidance three research scholars are awarded Ph.D degree. His area interest includes Database Technology, Cloud Computing, Block Chain, Network Security, Data Science, Artificial Intelligent and Machine Learning.