Terminology. International Journal of Theoretical and Applied Issues in Specialized Communication - Current Issue

Over the past decades, automatic term or terminology extraction (ATE), a natural language processing (NLP) task that aims to identify terms from specific domains by providing a list of candidate terms, has been challenging due to the strong influence of domain-specific differences on term definitions. Leveraging the advances of large-scale language models (LLMs), we propose LlamATE, a framework to verify the impact of domain specificity on ATE when using in-context learning prompts in open-sourced LLM-based chat models, namely Llama-2-Chat. We evaluate how well the LLM-based chat (e.g., using reinforcement learning with human feedback (RLHF)) models perform with different levels of domain-related information in the dominant language in NLP research (e.g., English) and other European languages (e.g., French, Slovene) from ACTER datasets, i.e., in-domain and cross-domain demonstrations with and without domain enunciation. Furthermore, we examine the potential of cross-lingual and cross-domain prompting to reduce the need for extensive data annotation of the target domain and language. The results demonstrate the potential of implicit in-domain learning where examples of the target domain are used as demonstrations for the prompts without specifying the domain of each example, and cross-lingual learning when knowledge is transferred from the dominant to lesser-represented European languages as for the data used to pre-train the LLMs. LlamATE also offers a valuable compromise by reducing the need for extensive data annotation, making it suitable for real-world applications where labeled corpora are scarce. The source code is publicly available at the following link: https://github.com/honghanhh/terminology2024.

Medical terminology is perceived as an obstacle for patients and family members to understand the medical message. In this context, plain language advocates for making specialised knowledge accessible to citizens. This article puts the synergy between medical terminology, plain language, and computational linguistics (a branch of Artificial Intelligence) on the table. Our purpose is to determine if Generative AI applications, like ChatGPT, can assist in creating glossaries of terms with their corresponding simple variants. For this, a relevant sub-field of medicine, cardiology, was taken as a case study, even though it could be extrapolated to other sub-fields. Next, a glossary of key cardiology terms was created following a classic methodological approach. Then, the most relevant phases of the process (terminology extraction, search for synonyms, and selection of the clearest synonym) were reproduced using ChatGPT. The results of the comparative study between both approaches give a glimpse of the extent to which this technology can be a useful resource for creating glossaries that can be employed for writing using plain language.

The interoperability of language resources is crucial for effective communication and data exchange across various computational systems. In this context, the ISO/TC 37 standards, specifically the Lexical Markup Framework (LMF) and the Terminological Markup Framework (TMF), play a vital role by providing a common framework for the modelling, representation, and exchange of lexical and terminological data. The LMF has been deliberately aligned with TMF to facilitate close coordination between the two standards. This paper explores the convergence between LMF and TMF, underscoring the need for a Unified Markup Framework (UMF) that enhances interoperability and effective resource management. We propose a unified meta-model that integrates these frameworks through comparative analysis and real-world examples, facilitating the development of advanced language processing applications and multilingual lexicographic and terminology management. This study not only underscores the opportunities and challenges of such coordination but also sets the groundwork for future research directions in the harmonisation of lexicographic and terminology resources.

A crucial task in any type of terminology work is identifying and extracting terms from relevant sources, which can be done manually or via (semi-)automatic term extraction processes. Given the recent advances in automatic term extraction (ATE) research, this paper explores the impact of ATE on terminology work in institutional settings (academic institutions, administrations, European institutions and international organizations) based on qualitative data. The analysis of 15 semi-structured expert interviews conducted in 2023 shows that the newest advances in research in ATE have not had an immediate impact on terminology practices in institutional settings for the study participants. This paper aims to discuss the reasons for the slow uptake of ATE in institutional settings, such as the gap between ATE tools developed in research and ATE components integrated in off-the-shelf terminology or corpus management systems, the lack of integration into existing workflows, the lack of support for certain languages, especially for less-resourced languages, as well as reasons related to source materials.

Terminology management is crucial in internationalised environments, particularly in higher education. While the international dimension has long been a part of academia, there is a scarcity of studies on terminology management in multilingual and multicultural settings. This study focuses on developing a normative-harmonising terminology management protocol for the Arqus European University Alliance. It also involves creating and implementing a prototype multilingual termbase aligned with the TBX standard for terminology data exchange. This resource supports the management protocol and incorporates functionalities identified in a preliminary user needs questionnaire (Montero-Martínez et al. 2020). The paper highlights the absence of terminology resources and language policies in academic institutions and underscores the challenges in establishing institutional termbases. The resulting prototype addresses these gaps, aiming to enhance institutional consistency and communication among the member universities.