The modification of the SBERT language model for identifying ESG risks based on textual data from companies and supervisory activities
Vestnik of Saint Petersburg University. Applied Mathematics. Computer Science. Control Processes · 2025 · 21 (1) · 75-91
DOI: 10.21638/spbu10.2025.106 ↗Kurz gefasst
Fachartikel in der Zeitschrift Vestnik of Saint Petersburg University. Applied Mathematics. Computer Science. Control Processes, 2025. Verfasst von Aleksey V. Buzmakov und 6 weiteren Autor:innen.
Abstract (Originalsprache)
An approach has been developed to identify risks associated with companies’ environmental impact, social responsibility, and governance quality (Environmental, Social, and Governance — ESG risks) based on textual information about the company. To achieve this, a modification of the SBERT language model is proposed with a clearly defined distance function for the embedding space. The model is trained on data from supervisory activities and texts of corporate websites. An example of interpretation of the model’s result is provided.
Zitieren
APA
Buzmakov, A. V., Kirpishchikov, D. A., Naidenova, Y. N., Paklina, S. N., Parshakov, P. A., Solomatin, R. I., & Sotiriadi, N. S. (2025). The modification of the SBERT language model for identifying ESG risks based on textual data from companies and supervisory activities. Vestnik of Saint Petersburg University. Applied Mathematics. Computer Science. Control Processes, 21(1), 75-91. https://doi.org/10.21638/spbu10.2025.106
Vancouver
Buzmakov AV, Kirpishchikov DA, Naidenova YN, Paklina SN, Parshakov PA, Solomatin RI, et al. The modification of the SBERT language model for identifying ESG risks based on textual data from companies and supervisory activities. Vestnik of Saint Petersburg University. Applied Mathematics. Computer Science. Control Processes. 2025;21(1):75-91. doi:10.21638/spbu10.2025.106
BibTeX
@article{Buzmakov2025,
author = {Buzmakov, Aleksey V. and Kirpishchikov, Dmitriy A. and Naidenova, Yuliya N. and Paklina, Sofiya N. and Parshakov, Petr A. and Solomatin, Roman I. and Sotiriadi, Nazar S.},
title = {The modification of the SBERT language model for identifying ESG risks based on textual data from companies and supervisory activities},
journal = {Vestnik of Saint Petersburg University. Applied Mathematics. Computer Science. Control Processes},
year = {2025},
volume = {21},
number = {1},
pages = {75--91},
doi = {10.21638/spbu10.2025.106}
}
EUVIMED