Training and analyzing a Transformer-based machine translation model
Autor(a) principal: | |
---|---|
Data de Publicação: | 2024 |
Outros Autores: | |
Tipo de documento: | Artigo |
Idioma: | por |
Título da fonte: | Texto livre |
Texto Completo: | https://periodicos.ufmg.br/index.php/textolivre/article/view/49118 |
Resumo: | The objective of this work is to analyze transformer-based machine translation models. It aims to test the feasibility of using trained models based on specialized corpus. For the training of such model, a parallel English-French corpus was built with seven texts related to the Convention of 25 October 1980 on the Civil Aspects of International Child Abduction. The translation results obtained by the trained model were compared with those produced by Google Translate. For the evaluation stage, sacreBLEU automatic evaluation and human evaluation methods were used. The outcome of the automatic evaluation of sentences produced by the trained model was, on average, higher than those generated by the non-trained model. The human evaluation of the sentences revealed that there were adequacy errors in the use of language specific to the subject matter of the 1980 Hague Convention both in sentences generated by the trained model and in sentences generated by the Google Translate model. |
id |
UFMG-9_b211a71cb5af562e19b85e59b1009c4d |
---|---|
oai_identifier_str |
oai:periodicos.ufmg.br:article/49118 |
network_acronym_str |
UFMG-9 |
network_name_str |
Texto livre |
repository_id_str |
|
spelling |
Training and analyzing a Transformer-based machine translation modelTreinamento e análise de um modelo de tradução automática baseado em TransformerLinguística computacionalTradutor automáticoTransformerCorpus paraleloAvaliação de tradução automáticaComputational LinguisticsMachine translationTransformerParallel corpusMachine translation evaluationThe objective of this work is to analyze transformer-based machine translation models. It aims to test the feasibility of using trained models based on specialized corpus. For the training of such model, a parallel English-French corpus was built with seven texts related to the Convention of 25 October 1980 on the Civil Aspects of International Child Abduction. The translation results obtained by the trained model were compared with those produced by Google Translate. For the evaluation stage, sacreBLEU automatic evaluation and human evaluation methods were used. The outcome of the automatic evaluation of sentences produced by the trained model was, on average, higher than those generated by the non-trained model. The human evaluation of the sentences revealed that there were adequacy errors in the use of language specific to the subject matter of the 1980 Hague Convention both in sentences generated by the trained model and in sentences generated by the Google Translate model.O presente trabalho possui como objetivo a análise dos modelos de tradução automática baseados em Transformer. Em específico, a pesquisa visa o teste da viabilidade do uso de modelos treinados a partir de corpus especializado. Para o treinamento do modelo, foi construído um corpus paralelo inglês-francês a partir de sete textos da Convenção de 25 de outubro de 1980 sobre os Aspectos Civis do Rapto Internacional de Crianças. Os resultados de tradução obtidos pelo modelo treinado foram comparados com aqueles produzidos pelo Google Tradutor. Para a etapa de avaliação foram utilizados os métodos de avaliação automática sacreBLEU e avaliação humana. Os resultados da avaliação automática de frases produzidas pelo modelo treinado foram, em média, mais positivos que aqueles gerados pelo modelo não treinado. A avaliação humana das frases revelou que houve erros de adequação no uso da linguagem específica à matéria da Convenção da Haia de 1980 tanto em frases geradas pelo modelo treinado, quanto em frases geradas pelo modelo do Google Tradutor.Universidade Federal de Minas Gerais2024-04-11info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionArtigo avaliado pelos paresapplication/pdfhttps://periodicos.ufmg.br/index.php/textolivre/article/view/4911810.1590/1983-3652.2024.49118Texto Livre; Vol. 17 (2024): Texto Livre: Linguagem e Tecnologia; e49118Texto Livre; Vol. 17 (2024): Texto Livre: Linguagem e Tecnologia; e49118Texto Livre; Vol. 17 (2024): Texto Livre: Linguagem e Tecnologia; e49118Texto Livre; v. 17 (2024): Texto Livre: Linguagem e Tecnologia; e491181983-3652reponame:Texto livreinstname:Universidade Federal de Minas Gerais (UFMG)instacron:UFMGporhttps://periodicos.ufmg.br/index.php/textolivre/article/view/49118/43842Copyright (c) 2024 Clóvis Henrique Martins Pimentel, Thiago Blanch Pireshttps://creativecommons.org/licenses/by/4.0info:eu-repo/semantics/openAccessPimentel, Clóvis Henrique MartinsPires, Thiago Blanch2024-05-12T20:28:03Zoai:periodicos.ufmg.br:article/49118Revistahttp://www.periodicos.letras.ufmg.br/index.php/textolivrePUBhttps://periodicos.ufmg.br/index.php/textolivre/oairevistatextolivre@letras.ufmg.br1983-36521983-3652opendoar:2024-05-12T20:28:03Texto livre - Universidade Federal de Minas Gerais (UFMG)false |
dc.title.none.fl_str_mv |
Training and analyzing a Transformer-based machine translation model Treinamento e análise de um modelo de tradução automática baseado em Transformer |
title |
Training and analyzing a Transformer-based machine translation model |
spellingShingle |
Training and analyzing a Transformer-based machine translation model Pimentel, Clóvis Henrique Martins Linguística computacional Tradutor automático Transformer Corpus paralelo Avaliação de tradução automática Computational Linguistics Machine translation Transformer Parallel corpus Machine translation evaluation |
title_short |
Training and analyzing a Transformer-based machine translation model |
title_full |
Training and analyzing a Transformer-based machine translation model |
title_fullStr |
Training and analyzing a Transformer-based machine translation model |
title_full_unstemmed |
Training and analyzing a Transformer-based machine translation model |
title_sort |
Training and analyzing a Transformer-based machine translation model |
author |
Pimentel, Clóvis Henrique Martins |
author_facet |
Pimentel, Clóvis Henrique Martins Pires, Thiago Blanch |
author_role |
author |
author2 |
Pires, Thiago Blanch |
author2_role |
author |
dc.contributor.author.fl_str_mv |
Pimentel, Clóvis Henrique Martins Pires, Thiago Blanch |
dc.subject.por.fl_str_mv |
Linguística computacional Tradutor automático Transformer Corpus paralelo Avaliação de tradução automática Computational Linguistics Machine translation Transformer Parallel corpus Machine translation evaluation |
topic |
Linguística computacional Tradutor automático Transformer Corpus paralelo Avaliação de tradução automática Computational Linguistics Machine translation Transformer Parallel corpus Machine translation evaluation |
description |
The objective of this work is to analyze transformer-based machine translation models. It aims to test the feasibility of using trained models based on specialized corpus. For the training of such model, a parallel English-French corpus was built with seven texts related to the Convention of 25 October 1980 on the Civil Aspects of International Child Abduction. The translation results obtained by the trained model were compared with those produced by Google Translate. For the evaluation stage, sacreBLEU automatic evaluation and human evaluation methods were used. The outcome of the automatic evaluation of sentences produced by the trained model was, on average, higher than those generated by the non-trained model. The human evaluation of the sentences revealed that there were adequacy errors in the use of language specific to the subject matter of the 1980 Hague Convention both in sentences generated by the trained model and in sentences generated by the Google Translate model. |
publishDate |
2024 |
dc.date.none.fl_str_mv |
2024-04-11 |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion Artigo avaliado pelos pares |
format |
article |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
https://periodicos.ufmg.br/index.php/textolivre/article/view/49118 10.1590/1983-3652.2024.49118 |
url |
https://periodicos.ufmg.br/index.php/textolivre/article/view/49118 |
identifier_str_mv |
10.1590/1983-3652.2024.49118 |
dc.language.iso.fl_str_mv |
por |
language |
por |
dc.relation.none.fl_str_mv |
https://periodicos.ufmg.br/index.php/textolivre/article/view/49118/43842 |
dc.rights.driver.fl_str_mv |
Copyright (c) 2024 Clóvis Henrique Martins Pimentel, Thiago Blanch Pires https://creativecommons.org/licenses/by/4.0 info:eu-repo/semantics/openAccess |
rights_invalid_str_mv |
Copyright (c) 2024 Clóvis Henrique Martins Pimentel, Thiago Blanch Pires https://creativecommons.org/licenses/by/4.0 |
eu_rights_str_mv |
openAccess |
dc.format.none.fl_str_mv |
application/pdf |
dc.publisher.none.fl_str_mv |
Universidade Federal de Minas Gerais |
publisher.none.fl_str_mv |
Universidade Federal de Minas Gerais |
dc.source.none.fl_str_mv |
Texto Livre; Vol. 17 (2024): Texto Livre: Linguagem e Tecnologia; e49118 Texto Livre; Vol. 17 (2024): Texto Livre: Linguagem e Tecnologia; e49118 Texto Livre; Vol. 17 (2024): Texto Livre: Linguagem e Tecnologia; e49118 Texto Livre; v. 17 (2024): Texto Livre: Linguagem e Tecnologia; e49118 1983-3652 reponame:Texto livre instname:Universidade Federal de Minas Gerais (UFMG) instacron:UFMG |
instname_str |
Universidade Federal de Minas Gerais (UFMG) |
instacron_str |
UFMG |
institution |
UFMG |
reponame_str |
Texto livre |
collection |
Texto livre |
repository.name.fl_str_mv |
Texto livre - Universidade Federal de Minas Gerais (UFMG) |
repository.mail.fl_str_mv |
revistatextolivre@letras.ufmg.br |
_version_ |
1799711141169987584 |