Corpus e sistemas de memória de tradução: a ferramenta Livedocs do MemoQ

• Autor(a) principal: Aleixo, Carolina Miranda
• Data de Publicação: 2020
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Repositório Institucional da UFU
• Texto Completo: https://repositorio.ufu.br/handle/123456789/32004; http://doi.org/10.14393/ufu.di.2020.354
• topic: Competência tradutória; Translation Competence; Corpus; Sistema de memória de tradução; Translation Memory System; MemoQ; LiveDocs; Terminologia; Terminology; CNPQ::LINGUISTICA, LETRAS E ARTES; CNPQ::LINGUISTICA, LETRAS E ARTES::LINGUISTICA::LINGUISTICA APLICADA; Linguística de corpus; Serviços de tradução

Translation requires different knowledge and skills, such as declarative and procedural knowledge. PACTE (2017) group called the set of these skills Translation Competence (TC). TC is divided into five sub-competences: bilingual, instrumental, extra-linguistic, knowledge about translation and strategic. In addition, PACTE (2017) also defines a group of psycho-physiological components, described as cognitive and psychomotor components. Based on bilingual and instrumental sub-competencies, this master's research analyses corpora as documentation and reference material in technical scientific translation by using a translation memory system (TMS). This research is unprecedented as it explores an emerging resource: the direct use of corpora in TMS. MemoQ is the TMS used in this study. Through the LiveDocs feature, it supports the use of corpora as documentation and reference material during translation. The domain used for research is Transactional Analysis (TA), a subarea of Psychology. The methodology is based on Corpus Linguistics. AntConc software was used in comparison with MemoQ to collect data for terminological analysis. Two bilingual corpora were compiled: a comparable corpus and a parallel corpus. The comparable corpus consists of a sub-corpus EO (original English) and a sub-corpus PO (original Portuguese). The parallel corpus consists of a parallel EO sub-corpus and a parallel PT (translated Portuguese) sub-corpus. In order to analyse the functionality of the TMS, a text should be used as a source text. A terminological analysis was carried out in four stages: (i) term candidates were extracted from the source text; (ii) these candidates were confirmed by using the corpus EO; (iii) suggestions of equivalent terms were found in the parallel corpus; (iv) the corpus PO provided information to validate the equivalent terms. As a conclusion, it is possible to integrate monolingual and parallel corpora in MemoQ, offering the translator advantages for their work. When integrating any type of corpora in MemoQ, the system outputs automatic searches during translation, showing results and combinations of words that are in the corpora and in the source text. As the final product of this research, a tutorial was created to demonstrate the use of corpora in MemoQ via the LiveDocs feature.