Business failure prediction from textual and tabular data with sentence-level interpretations
- Author
- Henri Arno (UGent) , Klaas Mulier (UGent) , Joke Baeck (UGent) and Thomas Demeester (UGent)
- Organization
- Project
- Abstract
- Business failure prediction models are crucial in high-stakes domains like banking, insurance, and investing. In this paper, we propose an interpretable model that combines numerical and sentence-level textual features through a well-known attention mechanism. Our model demonstrates competitive performance across various metrics, and the attention weights help identify sentences intuitively linked to business failure, offering a form of interpretability. Furthermore, our findings highlight the strength of traditional financial ratios for business failure prediction while textual data-particularly when represented as keywords-is mainly useful to correctly classify corporate disclosures where the possibility of failure is explicitly mentioned.
- Keywords
- Decision support systems, Business failure prediction, Natural language processing, Text analytics, FINANCIAL RATIOS, BANKRUPTCY PREDICTION, LEARNING-MODELS, DISCLOSURE, DISTRESS
Downloads
-
publisher version.pdf
- full text (Published version)
- |
- open access
- |
- |
- 669.25 KB
Citation
Please use this url to cite or link to this publication: http://hdl.handle.net/1854/LU-01JSP0JPGD6Z6MFE5BN15QQ4QT
- MLA
- Arno, Henri, et al. “Business Failure Prediction from Textual and Tabular Data with Sentence-Level Interpretations.” ANNALS OF OPERATIONS RESEARCH, vol. 353, no. 2, 2025, pp. 667–92, doi:10.1007/s10479-025-06574-z.
- APA
- Arno, H., Mulier, K., Baeck, J., & Demeester, T. (2025). Business failure prediction from textual and tabular data with sentence-level interpretations. ANNALS OF OPERATIONS RESEARCH, 353(2), 667–692. https://doi.org/10.1007/s10479-025-06574-z
- Chicago author-date
- Arno, Henri, Klaas Mulier, Joke Baeck, and Thomas Demeester. 2025. “Business Failure Prediction from Textual and Tabular Data with Sentence-Level Interpretations.” ANNALS OF OPERATIONS RESEARCH 353 (2): 667–92. https://doi.org/10.1007/s10479-025-06574-z.
- Chicago author-date (all authors)
- Arno, Henri, Klaas Mulier, Joke Baeck, and Thomas Demeester. 2025. “Business Failure Prediction from Textual and Tabular Data with Sentence-Level Interpretations.” ANNALS OF OPERATIONS RESEARCH 353 (2): 667–692. doi:10.1007/s10479-025-06574-z.
- Vancouver
- 1.Arno H, Mulier K, Baeck J, Demeester T. Business failure prediction from textual and tabular data with sentence-level interpretations. ANNALS OF OPERATIONS RESEARCH. 2025;353(2):667–92.
- IEEE
- [1]H. Arno, K. Mulier, J. Baeck, and T. Demeester, “Business failure prediction from textual and tabular data with sentence-level interpretations,” ANNALS OF OPERATIONS RESEARCH, vol. 353, no. 2, pp. 667–692, 2025.
@article{01JSP0JPGD6Z6MFE5BN15QQ4QT,
abstract = {{Business failure prediction models are crucial in high-stakes domains like banking, insurance, and investing. In this paper, we propose an interpretable model that combines numerical and sentence-level textual features through a well-known attention mechanism. Our model demonstrates competitive performance across various metrics, and the attention weights help identify sentences intuitively linked to business failure, offering a form of interpretability. Furthermore, our findings highlight the strength of traditional financial ratios for business failure prediction while textual data-particularly when represented as keywords-is mainly useful to correctly classify corporate disclosures where the possibility of failure is explicitly mentioned.}},
author = {{Arno, Henri and Mulier, Klaas and Baeck, Joke and Demeester, Thomas}},
issn = {{0254-5330}},
journal = {{ANNALS OF OPERATIONS RESEARCH}},
keywords = {{Decision support systems,Business failure prediction,Natural language processing,Text analytics,FINANCIAL RATIOS,BANKRUPTCY PREDICTION,LEARNING-MODELS,DISCLOSURE,DISTRESS}},
language = {{eng}},
number = {{2}},
pages = {{667--692}},
title = {{Business failure prediction from textual and tabular data with sentence-level interpretations}},
url = {{http://doi.org/10.1007/s10479-025-06574-z}},
volume = {{353}},
year = {{2025}},
}
- Altmetric
- View in Altmetric
- Web of Science
- Times cited: