EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

TC-HGAT: A Task-Conditioned Hierarchical Graph Attention Network for ESG Text Classification

Yasser Elouargui, Abdellatif Sassioui, Meriyem Chergui, Mohamed El Kamili, Mohammed Ouzzif

Applied Sciences · 09.11.2026 · 16 (21) · 11092

DOI: 10.3390/app162111092 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Applied Sciences, online erschienen am 09.11.2026. Verfasst von Yasser Elouargui und 4 weiteren Autor:innen. Die Arbeit nennt 37 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Classifying corporate environmental, social, and governance (ESG) disclosures is challenging because they combine technical terminology with subtle, domain-specific distinctions. Pretrained Transformers such as ClimateBERT perform well but require many parameters and substantial GPU memory. This paper introduces TC-HGAT, a lightweight hierarchical graph attention network with task-semantic conditioning. TC-HGAT uses frozen FastText-300D embeddings and builds, for each document, a graph of word nodes, virtual cluster nodes, and a task-semantic node obtained by mean-pooling the embeddings of a natural-language task description. On three ESG tasks (Claim, Specificity, and Commitment), TC-HGAT achieves macro-F1 scores of 94.6%, 77.8%, and 79.3%, with the highest mean scores on Claim and Commitment, although no difference from ClimateBERT is statistically significant. Most of the gain comes from the task-semantic node, which improves macro-F1 significantly on all three tasks (by 5.7, 2.4, and 4.5 points), whereas the cluster layer alone gives a modest improvement that is significant only on Claim. Random vectors in the same graph position, including a norm-matched one, do not reproduce this gain, indicating that it depends on the content of the task representation. TC-HGAT trains about 1.4 million parameters, far fewer than BERT-base and ClimateBERT, and needs substantially less peak GPU memory during training.

Zitieren

APA

Elouargui, Y., Sassioui, A., Chergui, M., Kamili, M. E., & Ouzzif, M. (2026). TC-HGAT: A Task-Conditioned Hierarchical Graph Attention Network for ESG Text Classification. Applied Sciences, 16(21), 11092. https://doi.org/10.3390/app162111092

Vancouver

Elouargui Y, Sassioui A, Chergui M, Kamili ME, Ouzzif M. TC-HGAT: A Task-Conditioned Hierarchical Graph Attention Network for ESG Text Classification. Applied Sciences. 2026;16(21):11092. doi:10.3390/app162111092

BibTeX
@article{Elouargui2026,
  author = {Elouargui, Yasser and Sassioui, Abdellatif and Chergui, Meriyem and Kamili, Mohamed El and Ouzzif, Mohammed},
  title = {TC-HGAT: A Task-Conditioned Hierarchical Graph Attention Network for ESG Text Classification},
  journal = {Applied Sciences},
  year = {2026},
  volume = {16},
  number = {21},
  pages = {11092},
  doi = {10.3390/app162111092}
}
Kostenlos starten →