A novel method for credit scoring based on feature transformation and ensemble model

Hongxiang Li; Ao Feng; Bin Lin; Houcheng Su; Zixi Liu; Xuliang Duan; Haibo Pu; Yifei Wang

doi:10.7717/peerj-cs.579

A novel method for credit scoring based on feature transformation and ensemble model

PeerJ Comput Sci. 2021 Jun 4:7:e579. doi: 10.7717/peerj-cs.579. eCollection 2021.

Authors

Hongxiang Li^#¹, Ao Feng^#¹, Bin Lin¹, Houcheng Su¹, Zixi Liu¹, Xuliang Duan¹, Haibo Pu¹, Yifei Wang²

Affiliations

¹ College of Information Engineering, Sichuan Agricultural University, Ya'an, Sichuan, China.
² Research Institute of Economics and Management, Southwest University of Finance & Economics, Chengdu, Sichuan, China.

^# Contributed equally.

Abstract

Credit scoring is a very critical task for banks and other financial institutions, and it has become an important evaluation metric to distinguish potential defaulting users. In this paper, we propose a credit score prediction method based on feature transformation and ensemble model, which is essentially a cascade approach. The feature transformation process consisting of boosting trees (BT) and auto-encoders (AE) is employed to replace manual feature engineering and to solve the data imbalance problem. For the classification process, this paper designs a heterogeneous ensemble model by weighting the factorization machine (FM) and deep neural networks (DNN), which can efficiently extract low-order intersections and high-order intersections. Comprehensive experiments were conducted on two standard datasets and the results demonstrate that the proposed approach outperforms existing credit scoring models in accuracy.

Keywords: AutoEncoder; Boosting tree; Credit scoring; Deep neural network; Factorization machine; Feature transformation.

Grants and funding

This work was supported by the Sichuan Province Department of Education (Grant NO. JG2018-348) and the Sichuan Agricultural University (Grant NO. X2017036). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.