O uso de n-gramas de classe semântica em um corpus de aprendiz

• Autor(a) principal: Gil, Cristina Borges
• Data de Publicação: 2024
• Tipo de documento: Tese
• Idioma: por
• Título da fonte: Biblioteca Digital de Teses e Dissertações da PUC_SP
• Texto Completo: https://repositorio.pucsp.br/jspui/handle/handle/42166
• topic: CNPQ::LINGUISTICA, LETRAS E ARTES::LINGUISTICA::LINGUISTICA APLICADA; Linguística de corpus; Linguística de corpus de aprendiz; Chavicidade; Análise multidimensional; N-gramas de classe semântica; Corpus linguistics; Learner corpus research; Keyness; Multidimensional analysis; Semantic class n-grams

The aim of this paper is to analyze the use of semantic class n-grams (BERBER SARDINHA, 2023) in the written and oral production of learners of English as a foreign language. With this research, we evaluated whether variation in the use of semantic class n-grams can be explained by the fact that the text is written or spoken, by the task assigned to the learner, by their level of proficiency, by their mother tongue, by their age or by the years studying the English language. The corpus used in this study was COREFL, whose acronym stands for Corpus of English as a Foreign Language. First, the corpus was tagged with USAS, a semantic tagger. Next, the semantic class n-grams were extracted and selected and their keyness calculated. A Factor Analysis was carried out on these variables, a standard procedure for Multidimensional Analysis, and the factors were interpreted. We observed that the task and the mode play an important role in the variation of the semantic class n-grams used by the learners