Analysis of Rule-Based Machine Translation and Neural Machine Translation Approaches for Translating Portuguese to LIBRAS

  • Caio César M. de Oliveira UFPB
  • Thaís G. do Rêgo UFPB
  • Manuella A. C. B. Lima UFPB
  • Tiago Maritan U. Araújo UFPB

Resumo


In this paper, we propose a rule-based machine translator for Brazilian Portuguese to Brazilian Sign Language translation. This translator was implemented using the part of speech tagging and lemmatization techniques with implementations in Aelius and CoGrOO, respectively. Then, we developed a convolutional translator seeking to replicate the rule-based translation with a sintetic corpus. In this corpus, preprocessing techniques such as substitution of names, numbers and spelling errors by symbols were applied to improve processing. The translator were tested on two corpus, Bosque e OpenSub (extracted from a site of subtitles), of 69 and 36.858 lines respectively, and compared with translations generated by interpreters and translations generated by the application VLibras (LAVID-UFPB).
Publicado
29/10/2019
Como Citar

Selecione um Formato
OLIVEIRA, Caio César M. de; RÊGO, Thaís G. do; LIMA, Manuella A. C. B.; ARAÚJO, Tiago Maritan U.. Analysis of Rule-Based Machine Translation and Neural Machine Translation Approaches for Translating Portuguese to LIBRAS. In: ANAIS PRINCIPAIS DO SIMPÓSIO BRASILEIRO DE SISTEMAS MULTIMÍDIA E WEB (WEBMEDIA), 25. , 2019, Rio de Janeiro. Anais Principais do XXV Simpósio Brasileiro de Multimídia e Web. Porto Alegre: Sociedade Brasileira de Computação, oct. 2019 . p. 117-124.

Artigos mais lidos do(s) mesmo(s) autor(es)