Advancing enterprise credit rating through multi-head self-attention and transfer learning
PeerJ Computer Science · 24.06.2026 · 12 · e3920
DOI: 10.7717/peerj-cs.3920 ↗Kurz gefasst
Fachartikel in der Zeitschrift PeerJ Computer Science, online erschienen am 24.06.2026. Verfasst von Rui Zhang und 5 weiteren Autor:innen. Die Arbeit nennt 54 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Science and Technology Research Program of Chongqing Municipal Education Commission, Open Research Project of Chongqing Engineering Research Center for Intelligent Applications of Financial Big Data, 2022 High-Level Talent Research Startup Fund of Chongqing Engineering College.
Abstract (Originalsprache)
This study introduces the multi-head self-attention rotary-positional encoding Informer (MHSA-RPEI), a novel enterprise credit rating model that leverages self-correlation weights and positional information of financial indicators to enhance rating accuracy. Corporate credit rating presents challenges such as temporal dynamics, complex non-linear relationships, and cross-regional variations, which traditional and deep learning models struggle to address. The MHSA-RPEI model overcomes these limitations through four innovations: (1) objective MHSA-RPEI model, which is specifically designed for credit rating, achieving high precision using only financial data and serving as an objective and validated tool; (2) revolutionary rotary positional encoding (RoPE), which enhances temporal modeling, improving dynamic risk assessment; (3) strategic transfer learning (TL), which facilitates cross-regional adaptations, mitigating data scarcity and domain variations; and (4) optimized informer architecture, which improves long-term risk forecasting based on extensive financial time series. Experiments on the U.S., Chinese, and Japanese financial data showed that MHSA-RPEI achieved 92.16% accuracy (U.S.), 90.32% (China), and 92.50% (Japan), with area under the receiver operating characteristic curve scores ranging from 88.47% to 91.95%, surpassing traditional and state-of-the-art models. Furthermore, TL experiments highlighted the exceptional adaptability of the MHSA-RPEI for cross-regional applications, yielding performance gains of 5–6%. Ablation studies confirmed that the RoPE and self-correlation weight modules were crucial to the effectiveness of the model.
Förderung
- Science and Technology Research Program of Chongqing Municipal Education Commission
- Open Research Project of Chongqing Engineering Research Center for Intelligent Applications of Financial Big Data
- 2022 High-Level Talent Research Startup Fund of Chongqing Engineering College
Zitieren
APA
Zhang, R., Li, Q., Chen, J., Li, J., Sakdanuphab, R., & Li, H. (2026). Advancing enterprise credit rating through multi-head self-attention and transfer learning. PeerJ Computer Science, 12, e3920. https://doi.org/10.7717/peerj-cs.3920
Vancouver
Zhang R, Li Q, Chen J, Li J, Sakdanuphab R, Li H. Advancing enterprise credit rating through multi-head self-attention and transfer learning. PeerJ Computer Science. 2026;12:e3920. doi:10.7717/peerj-cs.3920
BibTeX
@article{Zhang2026,
author = {Zhang, Rui and Li, Qing and Chen, Jiangchuan and Li, Jie and Sakdanuphab, Rachsak and Li, Hualin},
title = {Advancing enterprise credit rating through multi-head self-attention and transfer learning},
journal = {PeerJ Computer Science},
year = {2026},
volume = {12},
pages = {e3920},
doi = {10.7717/peerj-cs.3920}
}
EUVIMED