Evaluating EcoLexiCAT: a Terminology-Enhanced CAT Tool

Evaluating EcoLexiCAT: a Terminology-Enhanced CAT Tool Pilar León-Araúz, Arianne Reimerink University of Granada Department of Translation and Interpreting, c/Buensuceso, 11 18002 Granada (Spain) {pleon, arianne}@ugr.es Abstract EcoLexiCAT is a web-based tool for the terminology-enhanced translation of specialized environmental texts for the language combination English-Spanish-English. It uses the open source version of the web-based CAT tool MateCat and enriches a source text with information from: (1) EcoLexicon, a multimodal and multilingual terminological knowledge base on the environment (Faber et al., 2014; Faber et al., 2016); (2) BabelNet, an automatically constructed multilingual encyclopedic dictionary and semantic network (Navigli & Ponzetto, 2012); (3) Sketch Engine, the well-known corpus query system (Kilgarriff et al., 2004); (4) IATE, the multilingual glossary of the European Commission; and (4) other external resources (i.e. Wikipedia, Collins, Wordreference, Linguee, etc.) that can also be customized by the user. The tool was built with the aim of integrating terminology management – often considered complex and time-consuming – in the translation workflow of a CAT tool. In this paper, EcoLexiCAT is described along the procedure with which it was evaluated and the results of the evaluation. Keywords: computer-assisted translation; terminology management; specialized translation 1. Introduction Currently, computer-assisted tools (CAT) do not seamlessly integrate terminology management – often considered complex and time-consuming – into the translation workflow. Furthermore, most terminological resources do not take into account the real search behavior of translators (Tudhope et al., 2006; Durán Muñoz, 2012: 78). Most terminological modules in CAT tools do not go beyond a simple glossary of source and target terms and access to corpora is rarely, if ever, provided. This leads to an inevitable loss of translation quality and a waste of precious time. To fill this void, we developed EcoLexiCAT (León-Araúz et al., 2017), a terminology-enhanced CAT tool that provides easy access to domain-specific terminological knowledge in context. This application integrates different features of the professional translation workflow in a stand-alone interface where a source text is interactively enriched with terminological information (i.e. definitions, translations, images, compound terms, corpus access, etc.) from different external resources: (1) EcoLexicon, a multimodal and multilingual terminological knowledge base (TKB) on the environment (Faber et al., 2014: Faber et al., 2016); (2) BabelNet, an automatically constructed multilingual encyclopedic dictionary and semantic network (Navigli and Ponzetto, 2012); (3) Sketch Engine, the well-known corpus query system (Kilgarriff et al., 2004); (4) IATE, the multilingual glossary of the European Commission; and (5) other external resources (i.e. Wikipedia, Collins, Wordreference, Linguee, etc.) that can also be customized by the user. The motivations behind the integration of these resources lie in the needs and expectations of translators regarding terminology management. According to Durán Muñoz (2012: 82), translators consider that the most important ISO fields (ISO 12630: 1999) in the microstructure of terminological resources are the following: clear and concrete definitions, equivalents, derivatives and compounds, domain specification, examples, phraseological information, definition in both languages for bilingual resources and abbreviations and acronyms. Likewise, translators believe that terminological resources should be able to (1) permit exportability …

Content truncated to 500 words