Skip to main

Vocabulary of natural language processing

Search from vocabulary

Concept information

applied NLP > machine translation > data-driven machine translation > statistical machine translation

Término preferido

statistical machine translation  

Definición

  • Machine translation which is generated from a parallel training corpus by calculating the most likely equivalent of a source word/phrase/sentence in the target language. ( https://zenodo.org/record/5646896)

Etiquetas alternativas

  • SMT

Ejemplo

  • In addition the field of statistical machine translation is rapidly progressing and the quality of systems is getting better and better. (Och & Ney, 2004)
  • Most statistical machine translation employs log-linear models that interpolate feature function values obtained from various submodels such as phrase tables and language models (LMs). (Imamura & Sumita, 2016)
  • Statistical machine translation (SMT) represents a translation process as a noisy channel model that consists of a source-channel model and a language model of the target language. (Sumita, 2002)
  • The rapid development of statistical machine translation (SMT) that we have seen in recent years would not have been possible without automatic metrics for measuring SMT quality. (Joty, Guzmán, Màrquez & Nakov, 2014)

En otras lenguas

URI

http://data.loterre.fr/ark:/67375/8LP-T48BKJNW-4

Descargue este concepto:

RDF/XML TURTLE JSON-LD última modificación 29/4/24