Authors: Xin Wang; Kai Zong; Cuicui Luo
Addresses: School of Computer Science and Technology, University of Chinese Academy of Sciences, Beijing, China ' School of Mathematical Sciences, University of Chinese Academy of Sciences, Beijing, China ' International College, University of Chinese Academy of Sciences, Beijing, China
Abstract: As the global economic environment has become more complicated in recent years, more and more credit bonds have defaulted. The credit risk early warning model plays a very effective role in preventing and controlling financial risk and debt default. This paper uses machine learning methods to establish a credit default risk prediction framework. In this paper, the oversampling technique is first applied to deal with imbalanced credit default data sets and then the credit risk detection performance of several machine learning algorithms is compared. The empirical results show that the performance of the ensemble learning algorithms is the best.
Keywords: machine learning; credit risk detection; ensemble learning.
International Journal of Financial Services Management, 2022 Vol.11 No.3, pp.183 - 189
Received: 22 Nov 2021
Accepted: 24 Nov 2021
Published online: 09 Nov 2022 *