Word- and sentence-level confidence measures for machine translation

Sylvain Raybaud 1, * Caroline Lavecchia 1 David Langlois 1 Kamel Smaïli 1
* Corresponding author
1 PAROLE - Analysis, perception and recognition of speech
INRIA Lorraine, LORIA - Laboratoire Lorrain de Recherche en Informatique et ses Applications
Abstract : A machine translated sentence is seldom completely correct. Confidence measures are designed to detect incorrect words, phrases or sentences, or to provide an estimation of the probability of correctness. In this article we describe several word- and sentence-level confidence measures relying on different features: mutual information between words, n-gram and backward n-gram language models, and linguistic features. We also try different combination of these measures. Their accuracy is evaluated on a classification task. We achieve 17% error-rate (0.84 f-measure) on word-level and 31% error-rate (0.71 f-measure) on sentence-level.
Keywords : confidence measure machine translation mutual information
Document type :
Conference papers
Complete list of metadatas

Cited literature [19 references]  Display  Hide  Download
https://hal.inria.fr/inria-00417541
Contributor : Sylvain Raybaud <>
Submitted on : Wednesday, September 16, 2009 - 10:43:29 AM
Last modification on : Thursday, January 11, 2018 - 6:19:56 AM
Long-term archiving on: Tuesday, October 16, 2012 - 10:56:02 AM

File

CM-EAMT.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : inria-00417541, version 1

Collections

CNRS | INRIA | LORIA | UNIV-LORRAINE

Citation

Sylvain Raybaud, Caroline Lavecchia, David Langlois, Kamel Smaïli. Word- and sentence-level confidence measures for machine translation. 13th Annual Meeting of the European Association for Machine Translation - EAMT 09, May 2009, Barcelona, Spain. ⟨inria-00417541⟩

Export

BibTeX TEI DC DCterms EndNote

Share

Metrics

Record views

364

Files downloads

304

 

 

Contact

archive-ouverte@inria.fr