Os lexical bundles na busca por semelhanças em um documento do setor farmacêutico

• Autor(a) principal: Mazza, Luciene Novais
• Data de Publicação: 2009
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Biblioteca Digital de Teses e Dissertações da PUC_SP
• Texto Completo: https://tede2.pucsp.br/handle/handle/14101
• topic: Lingüística de corpus; Site Master File; Corpus (Linguistica); Lingua inglesa -- Analise do discurso; Lexical bundles; Corpus linguistics; CNPQ::LINGUISTICA, LETRAS E ARTES::LINGUISTICA::LINGUISTICA APLICADA

The present study explored a specific document of the pharmaceutical segment called Site Master File through the investigation of words combinations defined as lexical bundles (Biber et al.,1999). The aim of the study was to draw out the bundles so that to verify the degree of conformity of the linguistic features the use of lexical bundles may achieve, as being part of a document organized in a similar way, produced by different authors at different locations around the world. The theoretical-methodological approach was developed on the principles of Corpus Linguistics (Stubbs, 1996; Scott and Tribble, 2006; Berber Sardinha and Barbara, 2008; amongst others), an approach that makes use of a vast variety of authentic texts of language in use supported by computational tools. We compiled for this study fifteen samples of the Site Master File document stored in machine-readable form that belong to the same multinational pharmaceutical company based in Europe, which has more than a hundred of plants situated across the world. The Site Master File is a document prepared by pharmaceutical manufacturers that contains specific information about the quality assurance, the production and quality control of pharmaceutical manufacturing operations carried out at a named site/plant in order to be submitted to a regulatory authority. In addition, all documents must be officially certified in English. The analysis of the corpus data was performed to extract three-word bundles by using scripting languages such as Perl and Cygwin. Besides, a computer application was also designed to provide the cross-reference of data. The results of data analysis showed that although the samples of Site Master File bring a large range of similarity in its organization, we have not found regularity on the use of recurrent lexical bundles across the Site Master File documents. Thus, considering the absences of common lexical bundles across documents, we observed that, in each operating area of the pharmaceutical business unit there are some typical characteristics in relation to the type of product manufactured in the site, the processes engaged in the unit pharmaceutical operations as well as the geographic nearness relationships to the linguistic choices made by the different authors. Therefore, this study offers a contribution to the knowledge of variation in English use in preparing the Site Master File by authors allocated in a specific site. Moreover, the present study involves further research into the field of English for Specific Purposes based on corpora and into the studies of terminology