Skip to Main content Skip to Navigation
Master Thesis

Adaptation au domaine de la traduction automatique statistique

Abstract : Statistical machine translation performances rely heavily on the quality and quantity of available training data. However, most suitable parallel and freely available corpora are from international organizations (UN, EU, etc.). As a result, the "style" of this training data is not very suitable to build a translation system adapted to specific fields. We present, then, in this work several techniques and approaches to adapt a translation system to a different specific domain for which there is a reasonable amount of training data. We achieved significant improvements in BLEU score with translation systems "adapted" from English to French.
Document type :
Master Thesis
Complete list of metadata

Cited literature [21 references]  Display  Hide  Download

https://dumas.ccsd.cnrs.fr/dumas-00631003
Contributor : Uga - Bulles <>
Submitted on : Tuesday, October 11, 2011 - 11:46:30 AM
Last modification on : Sunday, March 14, 2021 - 9:54:03 PM
Long-term archiving on: : Thursday, January 12, 2012 - 2:26:28 AM

Identifiers

  • HAL Id : dumas-00631003, version 1

Citation

Marwen Azouzi. Adaptation au domaine de la traduction automatique statistique. Linguistique. 2011. ⟨dumas-00631003⟩

Share

Metrics

Record views

310

Files downloads

761