Word- and Sentence-level Confidence Measures for Machine Translation

  title={Word- and Sentence-level Confidence Measures for Machine Translation},
  author={Sylvain Raybaud and Caroline Lavecchia and David Langlois and Kamel Smaı̈li},
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 wordand 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… CONTINUE READING
Highly Cited
This paper has 28 citations. REVIEW CITATIONS


Publications citing this paper.
Showing 1-10 of 12 extracted citations

Confidence estimation for machine translation using context vectors

7'th International Symposium on Telecommunications (IST'2014) • 2014
View 4 Excerpts
Highly Influenced


Publications referenced by this paper.
Showing 1-10 of 19 references

Confidence Estimation for Machine Translation

COLING • 2004
View 3 Excerpts
Highly Influenced

Mesures de confiance en traitement automatique de la parole et applications

J. Mauclair
Ph.D. thesis, LIUM, Le Mans, France. • 2006
View 1 Excerpt

A comparative study on various confidence measures in large vocabulary speech recognition

2004 International Symposium on Chinese Spoken Language Processing • 2004
View 2 Excerpts