Enhancing Bank Credit Risk Management Using the C5.0 Decision Tree Algorithm

Authors

  • Qi Xin Management Information Systems, University of Pittsburgh, Pittsburgh, PA, USA
  • Runze Song Information System & Technology Data Analytics, California State University, CA, USA
  • Zeyu Wang Computer Science, University of Toronto, Toronto, Canada
  • Zeqiu Xu Computer Science, Carnegie Mellon University, CA, USA
  • Fanyi Zhao Computer Science, Stevens Institute of Technology, NJ, USA

Keywords:

Credit Risk Management; Decision Tree Model; Loan Default Prediction; Machine Learning in Finance

Abstract

The study explores the application of the C5.0 decision tree algorithm to improve bank credit risk management. Banks can enhance their credit risk management practices by transforming risk identification from subjective judgment to objective analysis, risk measurement from qualitative to quantitative, and risk control from static to dynamic. Using Center for Machine Learning and Intelligent Systems data, we constructed a C5.0 decision tree model to predict high-risk bank loans. The model's performance was evaluated through various metrics, including a confusion matrix, revealing an error rate of 14.9%. The study demonstrates that decision tree models can significantly enhance the accuracy and efficiency of bank credit risk assessments by leveraging key features such as checking and savings balances.

Author Biographies

Qi Xin, Management Information Systems, University of Pittsburgh, Pittsburgh, PA, USA

Management Information Systems, University of Pittsburgh, Pittsburgh, PA, USA

Runze Song, Information System & Technology Data Analytics, California State University, CA, USA

Information System & Technology Data Analytics, California State University, CA, USA

Zeyu Wang, Computer Science, University of Toronto, Toronto, Canada

Computer Science, University of Toronto, Toronto, Canada

Zeqiu Xu, Computer Science, Carnegie Mellon University, CA, USA

Computer Science, Carnegie Mellon University, CA, USA

Fanyi Zhao, Computer Science, Stevens Institute of Technology, NJ, USA

Computer Science, Stevens Institute of Technology, NJ, USA

Downloads

Published

2024-06-10