EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

TMCAS: Efficient Large Language Model-Assisted Topic Modeling for Civil Aviation Safety Reports

Xiangge Li, Haofeng Wang, Xiuting Zhou, Yan Ren, Zhi Tian, Weidong Liang, Gengsong Wang

Electronics · 05.08.2026 · 15 (15) · 3462

DOI: 10.3390/electronics15153462 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Electronics, online erschienen am 05.08.2026. Verfasst von Xiangge Li und 6 weiteren Autor:innen. Die Arbeit nennt 28 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Safety Capacity Project of the Civil Aviation Administration of China, Fundamental Research Funds of the China Academy of Civil Aviation Science and Technology.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Voluntary safety reports provide valuable information for identifying potential risks and improving safety management in civil aviation. However, these reports are often large in volume, unstructured in format, and rich in domain-specific terminology, making manual analysis costly, inefficient, and difficult to scale. To address these challenges, this paper proposes TMCAS, an efficient large language model-assisted topic modeling framework for civil aviation safety reports. The proposed framework combines domain-adapted text embeddings, density-based clustering, representative sampling, noise repair, and large language model-based topic generation. Specifically, a contrastive learning-based fine-tuning strategy is introduced to enhance the semantic representation of aviation safety texts. An HDBSCAN-based clustering and sampling mechanism is then designed to select representative reports and reduce the computational cost of large language model inference, while a noise-repair strategy is used to improve topic coverage. Finally, large language models are employed to generate interpretable sentence-level topic labels and descriptions. Experiments demonstrate that TMCAS achieves superior clustering and interpretability while substantially reducing inference cost compared with document-wise LLM baselines.

Förderung

  • Safety Capacity Project of the Civil Aviation Administration of China
  • Fundamental Research Funds of the China Academy of Civil Aviation Science and Technology

Zitieren

APA

Li, X., Wang, H., Zhou, X., Ren, Y., Tian, Z., Liang, W., & Wang, G. (2026). TMCAS: Efficient Large Language Model-Assisted Topic Modeling for Civil Aviation Safety Reports. Electronics, 15(15), 3462. https://doi.org/10.3390/electronics15153462

Vancouver

Li X, Wang H, Zhou X, Ren Y, Tian Z, Liang W, et al. TMCAS: Efficient Large Language Model-Assisted Topic Modeling for Civil Aviation Safety Reports. Electronics. 2026;15(15):3462. doi:10.3390/electronics15153462

BibTeX
@article{Li2026,
  author = {Li, Xiangge and Wang, Haofeng and Zhou, Xiuting and Ren, Yan and Tian, Zhi and Liang, Weidong and Wang, Gengsong},
  title = {TMCAS: Efficient Large Language Model-Assisted Topic Modeling for Civil Aviation Safety Reports},
  journal = {Electronics},
  year = {2026},
  volume = {15},
  number = {15},
  pages = {3462},
  doi = {10.3390/electronics15153462}
}
Kostenlos starten →