Efficient Combination of Confidence Measures for Machine Translation

Sylvain Raybaud 1, * David Langlois 1 Kamel Smaïli 1
* Auteur correspondant
1 PAROLE - Analysis, perception and recognition of speech
INRIA Lorraine, LORIA - Laboratoire Lorrain de Recherche en Informatique et ses Applications
Abstract : We present in this paper a twofold contribution to Confidence Measures for Machine Translation. First, in order to train and test confidence measures, we present a method to automatically build corpora containing realistic errors. Errors introduced into reference translation simulate classical machine translation errors (word deletion and word substitution), and are supervised by Wordnet. Second, we use SVM to combine original and classical confidence measures both at word- and sentence-level. We show that the obtained combination outperforms by 14% (absolute) our best single word-level confidence measure, and that combination of sentence-level confidence measures produces meaningful scores.
Type de document :
Communication dans un congrès
10th Annual Conference of the International Speech Communication Association - INTERSPEECH 2009, Sep 2009, Brighton, United Kingdom. 2009, Proceedings of INTERSPEECH 2009
Liste complète des métadonnées

Littérature citée [13 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/inria-00417546
Contributeur : Sylvain Raybaud <>
Soumis le : mercredi 16 septembre 2009 - 10:49:16
Dernière modification le : jeudi 11 janvier 2018 - 06:19:56
Document(s) archivé(s) le : mardi 16 octobre 2012 - 10:56:09

Fichier

interspeech09.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : inria-00417546, version 1

Collections

Citation

Sylvain Raybaud, David Langlois, Kamel Smaïli. Efficient Combination of Confidence Measures for Machine Translation. 10th Annual Conference of the International Speech Communication Association - INTERSPEECH 2009, Sep 2009, Brighton, United Kingdom. 2009, Proceedings of INTERSPEECH 2009. 〈inria-00417546〉

Partager

Métriques

Consultations de la notice

203

Téléchargements de fichiers

102