Day 1 INNOVATION 16 November
Moderator: Henrik Nilsson
Generative AI Makes Terminology Management More Critical
Author: Arle Lommel
Institution: CSA Research
As Generative AI caught on over the last year, many organizations have questioned why they should continue to manage terminology. The widespread perception was that large language models (LLMs) would take care of this task by drawing on relevant training data. However, with experience, it has become clear that clear, managed terminology is more important than ever and is vital for improving the use of LLMs. In this presentation, CSA Research Senior Analyst Arle Lommel will discuss how organizations can apply terminology to work with LLMs and how terminology can serve as a crucial link between them and knowledge management tools such as knowledge graphs
ChatGPT prompting in a university course on Terminology
Author: Barbara Heinisch
Institution: University of Vienna
Higher education has to adapt to a world characterized by digitalization and the increasing role of artificial intelligence. Among these, ChatGPT has a decisive impact on higher education, fostering debates at universities whether to allow ChatGPT for student’s assignments and exams or how to integrate it in university courses in a didactically meaningful way.
In university courses on terminology, technology has always played a crucial role, including automatic term extraction systems, terminology management and workflow systems as well as visualisation tools. ChatGPT and other artificial intelligence systems can become another element in the toolbox of terminologists and thus in university courses on terminology. Therefore, to foster the students’ digital competences and subject-specific knowledge, I present first insights into ChatGPT prompting in a university course on terminology. Based on trial-and-error learning, students were asked to enter prompts addressing various steps in terminology work as well as prompts helping them to acquire knowledge of terminology.
The results demonstrate that subject-specific knowledge of terminology in combination with detailed prompts deliver the most promising results for certain steps in terminology work when using ChatGPT. However, for certain steps, such as the visualization of terminological concept systems ChatGPT as text-generative artificial intelligence can only provide suggestions. For the actual graphical implementation image-generative artificial intelligence might be useful, but this still needs to be evaluated. Regarding the acquisition of knowledge of terminology by means of ChatGPT in a university course with students being mainly not familiar with terminology work, the prompting strategies used did not prove useful.
Overall, the use of ChatGPT in a university course on terminology showed that prompting strategies are crucial to obtain suitable results for certain aspects of terminology work. The impact of the use of ChatGPT on the students’ subject-specific knowledge and skills needs to be investigated in the future.
Terminology in the Semantic Web: Automatic Converters
Authors: Paula Diez-Ibarbia; Patricia Martín-Chozas; Elena Montiel-Ponsoda
Institution: Universidad Politécnica de Madrid
In the last years, publication of data on the web has experienced massive growth. Consequently, language technologies have received a lot of attention. One of the research lines in this area is the use of structured data in formats of the Semantic Web, particularly RDF. This format is characterized by the use of standardized vocabularies and ontologies (e.g., SKOS) that allow the representation of concepts and the relationships between them. However, generating resources in this format can be complex for users who are unfamiliar with this paradigm. To ease this process, we propose an automatic conversion service for terminology resources, which will be developed within the context of the national INESData project, an initiative for the creation of a safe data space for data and service sharing. However, there are two main challenges in this proposal. Firstly, the diversity of data types in linguistic resources of the same nature (e.g., terminologies) makes it difficult to convert them to RDF. Secondly, the absence of a standardized model that covers the different representation needs of each resource. Therefore, this research seeks to analyze the type of data included in the different terminology resources in order to create a thorough RDF conversion service as well as to contribute to the development of a vocabulary for terminology representation on the Semantic Web.
Automating Term Management Activities with AI: A Case Study of Generative AI for Metadata Generation and Intelligent Terminology Extraction
Authors: Roberto Silva; Marta Callava
Institution: Amazon Web Services
Terminology Program Managers at Amazon AWS Marketing Localization and Training Localization, respectively
The management of terminology in organizations can be a tedious and time-consuming task, requiring significant effort and resources from terminologists, linguists, and project managers. In recent years, there has been a growing interest in leveraging artificial intelligence (AI) to automate term management activities, freeing up time and resources for other critical activities.
In this presentation, we will showcase two case studies on how AI can be used to automate term management activities. In the first use case, we will discuss the use of Generative AI for metadata generation in different locales. We will share our experience of using ChatGPT, Amazon Bedrock and other similar tools to generate accurate and consistent metadata, such as definitions and context sentences, for over 2000 terms across multiple languages, and present the potential benefits of Generative AI, including significant reductions in time and resources. In the second use case, we will explore intelligent terminology extraction, an NLP technique that can automatically identify and extract domain-specific terms from text. We will share our experience of using different AI tools to extract relevant terms from a large corpus of texts in multiple languages, and how this can help organizations build and maintain term databases and glossaries more efficiently.
In conclusion, this presentation will provide insights and practical recommendations for organizations interested in leveraging AI to automate their term management activities. The case studies presented in this session demonstrate how AI can automate certain tasks to significantly reduce the time and resources needed, thus freeing up valuable resources for other added-value tasks.
Moderator: Elena Chiocchetti
Ontologies in relation to terminology and innovative aspects connected to ontology work in terminology
Author: Laura Giacomini
Institution: University of Innsbruck
Terminology and the medical domain in the 21st century: weaving a tapestry of knowledge
Author: Sara Carvalho
Institution: University Aveiro/NOVA CLUNL
In the dynamic landscape of 21st-century healthcare, the intersection of Terminology and the medical domain plays a critical role in shaping the evolution of knowledge representation, organisation, and sharing. This talk focuses on the intricate tapestry woven by the interconnectedness of terminology work and the medical field, exploring six pivotal notions that guide this relationship.
The talk starts by addressing concept orientation, unfolding the importance of a double-dimensional approach to Terminology, as well as the need to accommodate evolving conceptual frameworks. Ontologies provide the structural foundation, allowing us to operationalise the conceptual landscape of medical knowledge. Standardisation emerges as another essential pillar, fostering consistency and clarity in communication across diverse healthcare ecosystems.
The presentation also explores the multidimensionality inherent in medical terminology, recognising its role in capturing the complexity of healthcare scenarios. Interoperability becomes a crucial thread in this endeavour, weaving diverse systems into (an ideally) seamless fabric of shared understanding. Collaboration is an additional overarching theme, emphasising the collective - and patient-inclusive - effort required to advance medical terminology and healthcare knowledge.
A case study on endometriosis exemplifies the complex interweaving of these notions, demonstrating their practical implications. Prevailing trends and anticipated challenges will be addressed in the concluding remarks, ultimately aiming to underscore the synergies involving terminology work and the medical domain, while also emphasising their potential to foster theoretical and methodological advancements that may be applied to other subject fields
Une approche terminologique pour l’organisation, la représentation et le partage des connaissances sur le trouble dysmorphique corporel
Authors: Federica Vezzani; Rute Costa
Institutions: Università degli Studi di Padova; Universidade NOVA de Lisboa
The project we present in this talk intends to enhance the inter- and transdisciplinary potential of Terminology as an effective research area for the optimal conceptual and linguistic representations of specialized knowledge in the medical field. In particular, this project entails the study of the terminology of Body Dysmorphic Disorder (BDD) as a relatively under-/misdiagnosed medical condition whose incidence has grown in correlation to the COVID-19 pandemic (Quittkat et al. 2020). Despite its worldwide prevalence and considerable impact on social and private life, BDD lacks in a stable and shared conceptualization among experts and is still inconsistently represented, both from the conceptual and linguistic perspectives, in the existing terminological systems for health informatics. To bridge these gaps, we propose a complementary terminological approach (Santos and Costa 2015) aiming at contributing to the definition of stable conceptual and linguistic dimensions of BDD knowledge across different languages.
In this talk, we focus on the analysis of the conceptual evolution of BDD and the variation occurring at the linguistic level. Furthermore, we present how information about BDD is disseminated to and perceived by non-experts through the analysis of a corpus of mass-media articles and the results of a public survey. Finally, we present the requirements for the design and implementation of the first FAIR (Findable, Accessible, Interoperable, and Reusable) multilingual terminological resource (Vezzani 2022) focused on BDD allowing to enhance its knowledge representation, organization, sharing, and reuse. The resource will be conceived to meet the needs of 1) the healthcare professionals involved in the diagnosis and treatment of the disorder; 2) the general public, in particular patients and informal caregivers, in order to raise awareness towards the prevalence of the disorder; 3) the language professionals, such as translators and communication mediators in health services, to support the translation process and international communication; and 4) the researchers working in e-Health and medical information evaluation campaigns for the early detection of psychopathologies.
Quittkat, Hannah L., Rainer Düsing, Friederike-Johanna Holtmann, Ulrike Buhlmann, Jennifer Svaldi, and Silja Vocks. 2020. “Perceived Impact of COVID-19 Across Different Mental Disorders: A Study on Disorder-Specific Symptoms.” Frontiers in psychology 11: 586246.
Santos, Claudia, and Rute Costa. 2015. “Domain Specificity: Semasiological and Onomasiological Knowledge Representation.” In Handbook of Terminology, ed. by Kockaert, Hendrik J., and Frieda Steurs, 153-179. Amsterdam: John Benjamins.
Vezzani, Federica. 2022. Terminologie numérique: conception, représentation et gestion. Bern: Peter Lang International Academic Publishers.
Moderator: Niina Nissila
Testing translation students’ user experience:
Is there a future for terminology databases?
Authors: Ana Ostroški Anić; Martina Bajčić; Larisa Grčić; Ivanka Rajh
Institutions: Institute of Croatian Language and Linguistics; University of Rijeka, Faculty of Law; University of Zadar; University of Zagreb, Faculty of Humanities and Social Sciences
For decades, terminology databases have been considered to be the best resources for finding structured, informative, and validated domain terminology. Starting from the first term-oriented databases, across large multidomain termbanks such as Eurotermbank (Vasiļjevs, Gornostay & Skadiņa, 2011) and IATE, to concept-oriented knowledge bases (Faber & San Martin, 2011; L’Homme, Subirats & Robichaud, 2016) and linked terminology according to linguistic linked data (Cimiano et al., 2015), having terminology stored in a fixed, monitored and authorized resource has been deemed the best solution for creating accessible terminology.
New generations of translation and terminology students have been trained to exploit machine translation tools with integrated terminology, not to mention relying on online translation services, e.g. Google Translate and DeepL Translate, developed using impressive neural translation models. The question then arises: to what extent are today’s students familiar with terminology databases, and do they find them useful as sources of terminology for translation purposes?
We designed an online survey to test the user experience of translation studies students when searching for terms in two multidomain termbases, IATE and the Croatian termbase Struna, as well as in small, domain specific databases, i.e. TermFrame, the database of karst terminology, and AirFrame, the database of semantic frames of aviation. A pilot survey was first conducted to verify the survey design and the relevance of questions. Preliminary results showed that students find IATE lacking certain definitions and equivalents in all languages, while the TermFrame database was deemed the best based on its presentation of search results, ease of use, and appropriate complexity of information. The pilot study also revealed that some students hadn’t managed to use search options appropriately, leading us to conclude that an initial short presentation of each database prior to the survey would yield better results.
The survey is planned to be distributed among students of all philological groups at Croatian universities, both at the undergraduate and graduate levels. The results will provide valuable conclusions regarding the use of contemporary terminology databases, and suggest ways to improve them and integrate them into other tools or applications that provide language services for translators and other language experts.
Cimiano, P.; McCrae, J. P.; Rodríguez Doncel, V.; Gornostay, T.; Gómez Pérez, A.; Siemoneit, B.; Lagzdins, A. 2015. Linked Terminologies: Applying Linked Data Principles to Terminological Resources. In Proceedings of eLex: Electronic Lexicography in the 21 century.
Faber, P.; San Martin, A. 2011. Linking specialized knowledge and general knowledge in EcoLexicon. Actes de la conférence Terminologie & Ontologie: Théories et Applications (TOTh) 2011. Annency. 47–61.
L’ Homme, M.-C.; Subirats, C.; Robichaud, B. 2016. A Proposal for combining “general” and specialized frames. Proceedings of the 5th Workshop on Cognitive Aspects of the Lexicon (CogALex – V). The COLING 2016 Organizing Committee. Osaka. 156–165.
The terminology of current affairs.
How can we make the terminological project sustainable when the topic is still under debate?
Authors: Marta Grané; Marta Sabater; Sandra Cuadrado
Institution: TERMCAT, Centre de Terminologia
The practice of terminology work has tended to focus on compiling terms used in established areas of knowledge in the academic or industrial world, in an administrative field, or in a particular profession. In these cases, both the scope of the discipline and the documentary sources for the term extraction can be identified, thanks to the bibliography of that discipline and the consensus among the professionals that work within it.
But what happens when current news highlights a phenomenon or an activity around which public debate and popular discourse is generated? Rapidly, terms identified in this discourse go beyond the boundaries of the more traditional academic disciplines. Furthermore, since this discourse becomes integrated in the news of the moment, the time factor in the response is very important if it is intended to intervene with respect to the terminological forms used.
Based on three real examples, we will analyse the organisation of a project which seeks to provide a terminological response to a topic that has emerged in the current news:
- The terminology of climate emergency: this is inspired by the need to compile a set of terms, originating from various disciplines of knowledge, which currently have a very high profile in today's news, due to the global climate crisis that requires citizens and governments to adopt measures aimed at adapting to and mitigating climate change.
- The terminology of video-on-demand and streaming platforms: this is inspired by the Catalan government's wish to persuade the most powerful streaming platforms to include more content in Catalan (Netflix, Disney+, Prime Video, Filmin, etc.)
- The terminology of sexual consent: this is inspired by the processing of the Spanish government's Law on Sexual Freedom in 2022 and 2023, which had a significant impact and aroused considerable social debate.
The presentation will outline the challenges involved in these experiences and the strategies that have been followed to achieve some satisfactory results.
Terminology work in the context of conversational interfaces enabling natural language interaction also in Basque between industrial production machines and operators
Author: Begoña Arrate
The project EKIN and its subsequent phase BERREKIN (currently in progress) are making advancements in the development and implementation of “conversational interfaces” as an interaction mechanism between operators and machines in the industrial production plants of the Basque Country, with the aim of facilitating and improving the productivity of certain processes that take place in this setting.
The project EKIN used the most advanced Artificial Intelligence technologies:
- To facilitate the development of conversational interfaces enabling natural language interaction between operators and industrial production machines.
- To optimize the neural models of speech recognition and synthesis (so that they can be embedded in electronic devices in the machines themselves) and automatisms to avoid the problems of privacy, security and latency of their deployments in the cloud.
To be able to develop the conversational interfaces also in Basque language, a corpus was compiled (which contained a series of manuals, dossiers, maintenance notes and human-machine interaction expressions in the reference languages, Basque and Spanish) and terminology work was carried out in order to harmonize the terms used in this setting.
EKIN defined three environments to be used as use cases for technology development:
- The programming of numerical controls
- The manufacturing and assembly processes
- Industrial maintenance
The aim of the current phase (BERREKIN) is to focus in a specific manufacturing and assembly process (profiling a workpiece) to carry out the real implementation and to proceed to validate the results before transferring and applying this technology in other industrial production plants. The terminology work that is being carried out in this new phase bases on the above-mentioned corpus, but oral expressions and terms used by operators will also be analysed in this phase, and suggestions for improvement will be made.
 Carried out during 2020 and 2021 by a consortium made up by reference technological centres from the Basque Country (Vicomtech and Tekniker), the company IKOR, the Advanced Centre in Manufacturing IMH, the University of the Basque Country and UZEI (the Basque Centre for Terminology and Lexicography) and Funded by SPRI (the Basque Government’s entity for promoting the Basque industry).
Terminologue as a teaching resource for terminology - assessing the user experience
Authors: Úna Bhreathnach; Gearóid Ó Cleircín
Institution: Dublin City University
In this paper, we share our experience of teaching undergraduate students the principles of terminology using Terminologue. Terminologue (terminologue.org) is a cloud-based, open-source terminology management tool developed by us and available in 15 languages. Terminologue was initially developed to manage the National Terminology Database for Irish (téarma.ie). As a publicly funded project it was decided to make the resource openly available.
We have noticed since its launch that much of Terminologue’s use has been in the educational domain and certainly this has been the case in our own university where it is used as a teaching tool on several modules. We decided to investigate the ways in which Terminologue is being used by educators and students, to find out how it meets their needs and to identify potential improvements. This may also give a broader perspective into the teaching of terminology in a wider context.
Day 2 SUSTAINABILITY 17 November
Moderator: Ana Ostroški Anić
ROUNDTABLE: Learning from the past applying to the future
Maja Bratanić; Teresa Cabré; Susanne Lervad; Fidelma Ní Ghallchobhair; Delyth Prys
Moderator: Sandra Cuadrado
Guide on Equity, Diversity and Inclusion Terminology & Inclusive Writing Guidelines and Resources
Author: Chantal Reid
Institution: Government of Canada’s Translation Bureau
Grassroots networks and communities have opened conversations, often reliving their own personal traumas, in an effort to increase our collective awareness. In 2021, the Clerk of the Privy Council published a Call to Action on Anti-Racism, Equity, and Inclusion in the Federal Public Service, stating among others that building a diverse, equitable and inclusive Public Service is both an obligation and an opportunity for the Canadian Government, aiming to put the full capacity of our entire pool of talent at the service of the Canadian population. The linguistic tools described below help provide a unified frame of reference for federal departments and agencies to further the conversation and build paths forward.
The Interdepartmental Terminology Committee on Equity, Diversity and Inclusion was established to co-develop a guide on the key terms and concepts. The main objective of this evergreen bilingual (English-French) guide is to help improve understanding of concepts related to equity, diversity, accessibility and inclusion. To meet this objective while ensuring that the perspectives of equity-denied groups were acknowledged and taken into account, a very rigorous process was put into place. This process helped navigate the sensitive nature of the terminology under study and facilitate its adoption.
In parallel, an interdepartmental working group established inclusive writing guidelines and resources, including practical tools in French and English, called l'Inclusionnaire and the Inclusionary, respectively, which provide concrete solutions to be applied in an inclusive writing context.
The proposed presentation would give an overview of the context under which both interdepartmental groups were established, delve into the processes adopted and the pitfalls avoided, and quickly show the end result.
To gender or not to gender, that is the question: gender-inclusive language in the legal context
Authors: Elisabeth Evers; Natascia Ralli
Institutions: Formerly TH Köln; Eurac Research, Institute for Applied Linguistics
How do terminology science and gender meet? Actually, they usually don’t. Terminology science deals with the study of concepts and their representations in specific domains or subjects (Drewer/Schmitz 2017:5-6). In this regard, gender is considered a property and not a characteristic (Winter 2021: 29). However, the current debate on gender-inclusive agentives1, i.e. linguistic forms that indicate an agent, such as job titles, professions, educational qualifications, functions, and roles (Bengochea 2017:200) calls for a reflection on how to store agentives within terminology.
Related to that, the following aspects need to be taken into consideration (Evers 2021, Ralli/De Camillis 2022):
- identification of input models for the lemmatization of agentives according to whether gender is seen as a property or a characteristic of the concept;
- types of agentives and degree of syntagmatic complexity, comprising complex terms where the agentives are objects;
- symmetrical representation of sexes (Begonchea 2017:216);
- drafting of inclusive definitions.
Based on investigations we conducted in the last three years, we will describe the complexity of this topic in our paper, while focusing on the classification of agentives and on the strategies to represent them in terminology databases by means of examples in German from the legal context.
This specifically means, that at first, we will identify different term types focusing on the different possibilities to present gender-inclusive designations in terminology databases. Furthermore, we will classify simple and complex terms referring to persons in German as well as compound nouns and discuss in which cases it might be best to use the nomen agentis as the designation. Finally, we will evaluate different approaches to represent binary and non-binary agentives in terminology databases using sample entries.
1 With this expression, we refer to women, men and genderqueer people.
Bengoechea M. (2017): “Agentives for women, a gap still to fill: dismissing non-sexist language policies in terminological resources”. In Paola Faini (Hrsg.), Terminological approaches in the European context, 200–221. Cambridge: Cambridge Scholars Publishing.
Drewer P, and Schmitz K-D. 2017, Terminologiemanagement. Grundlagen - Methoden - Werkzeuge (Kommunikation und Medienmanagement). Berlin: Springer Vieweg.
Evers E. 2021, Gendergerechte Sprache. Auswirkungen auf die Terminologielehre, Terminologiearbeit und Terminographie. Masterarbeit, TH Köln.
Evers E. (2022), “Geschlechtersensible Sprache – Wie bringe ich es meiner Terminologiedatenbank bei?”. In edition 2/22: 13-20.
RaDT-Arbeitspapier, Gender und Terminologie. Publication in preparation.
Ralli N, and De Camillis F. 2022: “How to integrate feminine agentives in terminological databases: a methodological reflection”. LSP 2022 - 23rd International Conference on Languages for Specific Purposes. September 12-13, 2022, NOVA University Lisbon.
Winter T. 2021, “Das Gendern. Terminologisch betrachtet ein Missverständnis mit Skandalpotential”. In edition 2/21: 29-30.
Gender bias in machine translation and what terminologists can do about it
Author: Michal Měchura
In machine translation, gender bias occurs when the machine has made an unjustified assumption about someone’s gender, for example if it has translated a gender-neutral source word into a stereotypically gendered target word (‘pilot’ → ‘male pilot’).
In this presentation, I will introduce a technique for solving the problem called gender rewriting. A gender rewriter is a tool which takes the output of machine translation and rewrites it into a different gender or into a gender-neutral neoform, based on instructions from a human user. I will demonstrate several existing gender rewriters, including my own project Fairslator (www.fairslator.com).
In the second half of the presentation I will argue that gender rewriters are heavily dependent on human-edited lexicons of gender-sensitive terms, and I will invite the audience to contribute to a spin-off project from Fairslator, an open-source multilingual database of gender-sensitive terms called Genderbase (www.terminologue.org/genderbase).
Moderator: Úna Breathnach
Building a bulwark against domain loss through cross-sectoral language policy – successes and remaining challenges of a work in progress
Author: Åse Wetås
Institution: Språkrådet Norway
Onomasiological approach in terminology: giving it up comes at a price
Author: Anca-Marina Velicu
Institution: University of Bucharest
One (potential) consequence of giving up the onomasiological approach– a strong tendency in contemporary terminology theories, those last 20 years or so – is the way neural machine translation (NMT) software like DeepL translates in English Wüsterian (in)famous texts such as:
- „Jede Terminologiearbeit geht von den Begriffen aus. Sie zielt auf scharfe Abgrenzung zwischen den Begriffen“(Wüster 1979: 1, §0.211, „Ausgehen von den Begriffen“).
DeepL proposes in all three cases terms as interlingual equivalent for Begriffen (it indicates concepts only as second variant, upon specific search for better formulations).
We found the same relative confusion in the answers AI Bing provided to questions about what should be defined, the term “dog”, or the corresponding concept.
Our hypothesis is that, since this obviously can’t have anything to do with general usage of the noun concept as described in general language monolingual dictionaries of English, it must be linked to a specific usage of term(s) instead of concept(s), in terminology work:
- “The dictionary gathers about a hundred terms related to mobile devices, in particular smartphones and tablets, with their definitions in Catalan (…).” (Presentation of a TERMCAT Dictionary, our emphasis)
Considering the role of the resources both the NMT and the AI assistants may exploit, we have compiled a corpus illustrating the abandoning of clearcut distinctions between level of objects, level of concepts and level of words/terms, in French for special purposes (including research papers or book reviews in terminology, and translation memories). Some of those text samples exhibit borderline semantic violations amounting to a complete deletion of the use-mention divide:
- ??Le cheval fiscal est un terme essentiellement administratif. (OK « Cheval fiscal » est un terme …)
Should those documented usages be treated as a kind of new “Ist-Norm” (Wüster 1979) and exploited in favor of anti-wüsterian rhetoric and theories, or rather should they be dismissed as irrelevant (since by hypothesis accidental) mistakes (vs errors – cf. Corder 1967: 167)?
We will try to suggest specific arguments in favor of the latter, while also evoking some (textual) proofs (mostly from Wüster 1968, 1979) against the idea that onomasiology is the mother of all evils, in terminological endeavors: concept-driven terminology isn’t exclusive neither of figurative meanings and polysemy (only systematic polysemy is problematic with terms vs words), nor of certain kinds of synonymy– contrary to Temmerman 2007: 29.
Even language variation may be studied “starting from concepts” in the framework of cognitive diachronic onomasiology (Blank 1999, 2000).
Blank, Andreas/Koch, Peter (1999b): “Onomasiologie et étymologie cognitive: l’exemple de la TÊTE”. In: Vilela, Mário/Silva Fátima (eds.): Atas do 1o Encontro de Linguística Cognitiva. Porto, 49-71.
Blank, Andreas/Koch, Peter (2000): “La conceptualisation du corps humain et la lexicologie diachronique romane”. In: Dupuy-Engelhardt, Hiltraud/Montibus, Marie-Jeanne (eds.): La lexicalisation des structures conceptuelles. Reims, 43-62.
Nunberg, Geoffrey, Zaenen, Annie (1997) :« La polysémie systématique dans la description lexicale », Langue française, n°113, Aux sources de la polysémie nominale, p. 12-23
doi : https://doi.org/10.3406/lfr.1997.5366
Temmerman, Rita (2007): "Approaches to terminology. Now that the dust has settled". In: Simonnæs, Ingrid (ed.). SYNAPS 20. Bergen: NHH. 27-36.
Wüster, Eugen (1968): The Machine Tool. An Interlingual Dictionary of Basic Concepts comprising An Alphabetical Dictionary and A Classified Vocabulary with Definitions and Illustrations. London: Technical Press.
Wüster, Eugen (1979, posth.): Einführung in die allgemeine Terminologielehre und terminologische Lexicographie, Wien: Springer.
Coherence Creates sustainability -Terminology work in the Nordic Countries
Authors: Anita Nuopponen; Henrik Nilsson
Institutions: University of Vasa; CAG Consoden
Presentation of the volume Terminology: Cognition, language and communication
Teresa Cabré; Mercè Lorente
This volume brings together a selection of M. Teresa Cabré's articles on terminology published after
1999 in journals of diverse nature and scope, many of which are difficult to access; articles in
languages other than English are here provided in English translation. As a whole, these articles aim
to represent the author's groundbreaking work on terminology, both from a theoretical as from a
methodological and applied point of view. Part I includes texts on three fundamental aspects of
terminology as a field of knowledge: Firstly, general articles on the rethinking of proposals made by
other authors and on the bases for the formulation of the Communicative Theory of Terminology
(CU). Secondly, articles that deal with the rethinking of the framework of this subject, with emphasis
on specialised languages and communication. And thirdly, on the object of study: the terminological
unit. Part II includes articles on methodology, international standards, and teaching terminology, and
texts that deal with the intersection of terminology with other fields: Documentation, Translation,
Neology, and Language Policy.
Related Topics: Cognition and language; Communication Studies; Terminology
Moderator: Begoña Arrate
Investigating domain loss: a contrastive study of Italian and Catalan in AI terminology
Authors: Maria Carmen Staiano; Marta Grané
Institutions: Università degli Studi di Macerata; TERMCAT
The influence of dominant languages and cultures on local linguistic ecosystems has been a subject of extensive debate across various professional domains. This debate often focuses on the risk of domain loss by both national and minority languages, particularly within professional fields where dominant languages serve as international lingua francas.
A noteworthy example can be observed in Italy, where the widespread adoption of English terms in digital communication has become prevalent. Unlike Spain, Italy lacks a robust normative language policy, leading to the extensive use of anglicisms that often remain unassimilated into the Italian language.
To mitigate the risk of domain loss and minimize the influence of dominant languages, numerous efforts have been made over the years to enrich languages with domain-specific terms. TERMCAT, in particular, has made this a primary objective, working towards the development and integration of terminology resources in the Catalan language. The need to create new and up-to-date language equivalents is particularly evident in certain domains, such as artificial intelligence.
To address these concerns, a contrastive analysis of the Italian and Catalan languages was conducted, specifically examining terminology related to artificial intelligence and natural language processing and its usage in informative texts.
Through this comprehensive investigation, the study sheds light on the challenges that arise from domain loss and explores the linguistic strategies employed, in both Catalan and Italian, to tackle them. The presentation emphasizes the implications of relying on loanwords and highlights the necessity of developing robust, domain-specific terminology, suggesting effective alternatives to anglicisms.
By analyzing the domain-specific linguistic landscape, the presentation aims to contribute to a holistic understanding of how dominant languages, such as English, impact linguistic diversity and the measures required to safeguard and promote local languages in specialised fields of communication.
Terminology in the domain of seafood: A comparative analysis Germany- Spain (ontologies)
Author: Irene Jimenez Alonso
Institution: University of Innsbruck
In the last few decades, the study of terminology has undergone a cognitive shift that has led to the development of several approaches that study the social, linguistic, and cognitive dimension of terms, such as Communicative Theory of Terminology (CTT) (Cabré, 1993) and Frame-Based Terminology (FBT) (Faber, 2009).The aim of this paper is to describe, represent, and compare the terminological reality in the domain of seafood from an economic-culinary perspective in Germany and Spain on the basis of CTT and FBT.
CTT was developed in the early 1990s and argues that the study of terminology should be based on a communicative perspective, taking into account aspects such as the communicators and the context of communication. The development of FBT started in 2007. It uses certain aspects of Frame Semantics (Fillmore, 1976) to conceptualise specialised domains and create non-language-specific representations through the analysis of domain events and through the study of the behaviour of terminological units in texts. Both theories share many of the same premises and propose the representation of the concepts of a domain in an ontology. FBT also proposes a representation in frames.
Ontologies and frames serve different functions. Ontologies provide a detailed representation of the domain and are intended for experts, while frames give an overview usable for non-experts. On the basis of this formalism, ontologies appear to be better suited to represent the taxonomy of a domain, while frames seem more useful for representing processes. FBT proposes to combine both in terminology work. Since the chosen domain is strongly taxonomic in nature, my focus in the representation was on the ontology. However, I also represented some central processes in a frame in order to give an accessible first overview and to see to what extent this representation works in a non-process-oriented domain.
Are terms in academic research appropriate? Perceptions and opinions of Estonian doctoral students
Authors: Peep Nemvalts; Helena Lemendik
Institutions: Tallinn University; University of Tartu and Tallinn University
Often choosing a language for scientific communication is not straightforward as academics, and students indeed feel increasing pressure to use English instead of any other national language, given the culture of internationalisation of higher education. In 2011–2012 our team of linguists and sociologists from Tallinn University accomplished the study "Estonian as a language of higher education and academic research" supported by the EU. As a part of this, in 2012, the team surveyed Estonian doctoral students' perceptions and attitudes to the usage of academic languages (Roosmaa et al. 2014).
As recent national education, research and language strategies for 2021–2035 in Estonia again aim to advance Estonian-medium higher education, we conducted a re-survey in 2022 (Nemvalts et al., 2023). The amount of doctoral students has diminished in ten years. However, the sample was representative both for 2012 and 2022, covering all the universities in Estonia. This paper presents two studies' key results and implications, discusses the current academic language situation and policies, and concentrates on terminological issues. As a part of the study, we focused on two aspects of the usage of terms by doctoral students:
● How do they assess their possibilities to find appropriate terms in the current academic environment?
● How do they perceive the appropriateness of terms they use in their research?
We compared two groups of doctoral students: those of social sciences and culture, incl. humanities, and those of natural sciences and technical disciplines. The paper will present the results and discuss some comments of the respondents.
Roosmaa, E-L.; Roosalu, T. & Nemvalts, P. 2014. Doktorantide teadustöö keele valikutest. Ülikool ja keelevahetus. Tartu Ülikooli ajaloo küsimusi XLII, 37-52. https://ojs.utlib.ee/index.php/TYAK/article/view/11876
Nemvalts, P.; Roosalu, T.; Roosmaa, E-L. & Lemendik, H. 2023 (to be published). Globalisation vs diversity in national languages in HE context: Case of Estonia. In: Proceedings of the 9th International Conference on Higher Education Advances. Editorial Universitat Politecnica de València.
Relationship between translators’ research behaviour and the overall target text quality in human translation and post-editing of machine translation
Authors: Márta Lesznyák; Eszter Sermann
Institution: University of Szeged, Faculty of Humanities and Social Sciences
Translation-related activities (TRA) are reflected in various translation competence models (PACTE 2003, TransCert 2013, eTransFair 2016, EMT 2022) and industry standards, such as ISO 17100. Translation scholars have long acknowledged TRA activities as crucial elements in the translation process and a fundamental aspect of translation competence (e. g. Enriquez Raído 2013, Hvelplund 2017, Massey & Ehrensberger-Dow 2011).
In the spring of 2020, the Translation and MT Post-Editing Competence Research Group of the University of Szeged, Hungary started a comprehensive study in order to reveal what role PACTE’s translation sub-competences play in human translation and in post-editing of machine translation. In this presentation, we analyse data collected in the spring and fall of 2022, when 12 first-year and 10 second-year master’s students of translation translated and post-edited a part of a copyright agreement from English into Hungarian. Based on the screen videos of students' workflow (recorded with the OBS Studio software), in the first phase of the analysis, we identified the online translation resources, and then we compared (a) the nature of resources used by translators and post-editors, on the one hand, and (b) the nature of resources used by first-year and second-year students, on the other hand (Sermann 2023). This presentation reports on the second phase of the study, in which 10 key terms (belonging to the legal and the academic domains) were selected from the source text, their equivalents in the target texts were evaluated, and finally, research directions and strategies were compared based on Gough (2019) study.
In the presentation we will focus on how students’ research behaviour is related to the success of finding the target equivalents of the key terms and to the overall quality of the target texts. Human translators’ and post-editors’ research behaviour will be compared, too, just as first year and second-year students research strategies.