Learning Implicit Editorial Rules for Medicinal Terminology Naming Using Rule-Compositional Cascading Prompts
Studies in Health Technology and Informatics · 15.07.2026
DOI: 10.3233/shti260963 ↗Kurz gefasst
Fachartikel in der Zeitschrift Studies in Health Technology and Informatics, erschienen am 15.07.2026. Verfasst von Hoa Ngo und 2 weiteren Autor:innen. Beteiligte Einrichtung: The Australian e-Health Research Centre, CSIRO. Hauptthemen laut MEDLINE: Terminology as Topic, Vocabulary, Controlled, Natural Language Processing, Editorial Policies. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).
Abstract (Originalsprache)
Medicinal terminologies depend on expert-authored fully specified names (FSNs) and preferred terms (PTs) that encode implicit editorial rules not formally represented in ontologies. This work explores the use of large language models to learn and apply these rules for automated medicinal terminology naming. Using existing OWL-defined concepts from the HSE Ireland medicinal terminology, this project evaluated a four-level approach ranging from direct fine-tuning to rule-compositional cascading prompts. Results across six medicinal concept types show that the cascading prompt framework further improves accuracy, numerical robustness, and interpretability by explicitly modelling editorial decision steps, enabling scalable and consistent FSN and PT generation.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- The Australian e-Health Research Centre, CSIRO.
Zitieren
APA
Ngo, H., Cordell, M., & McMurtrie, D. (2026). Learning Implicit Editorial Rules for Medicinal Terminology Naming Using Rule-Compositional Cascading Prompts. Studies in Health Technology and Informatics. https://doi.org/10.3233/shti260963
Vancouver
Ngo H, Cordell M, McMurtrie D. Learning Implicit Editorial Rules for Medicinal Terminology Naming Using Rule-Compositional Cascading Prompts. Studies in Health Technology and Informatics. 2026. doi:10.3233/shti260963
BibTeX
@article{Ngo2026,
author = {Ngo, Hoa and Cordell, Matt and McMurtrie, Dion},
title = {Learning Implicit Editorial Rules for Medicinal Terminology Naming Using Rule-Compositional Cascading Prompts},
journal = {Studies in Health Technology and Informatics},
year = {2026},
doi = {10.3233/shti260963}
}
EUVIMED