Phrase-level segmentation and labelling of machine translation errors

Frédéric Blain, Varvara Logacheva, Lucia Specia

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

6 Citations (Scopus)

Abstract

This paper presents our work towards a novel approach for Quality Estimation (QE) of machine translation based on sequences of adjacent words, the so-called phrases. This new level of QE aims to provide a natural balance between QE at word and sentence-level, which are either too fine grained or too coarse levels for some applications. However, phrase-level QE implies an intrinsic challenge: how to segment a machine translation into sequence of words (contiguous or not) that represent an error. We discuss three possible segmentation strategies to automatically extract erroneous phrases. We evaluate these strategies against annotations at phrase-level produced by humans, using a new dataset collected for this purpose.

Original languageEnglish
Title of host publicationProceedings of the 10th International Conference on Language Resources and Evaluation, LREC 2016
EditorsNicoletta Calzolari, Khalid Choukri, Helene Mazo, Asuncion Moreno, Thierry Declerck, Sara Goggi, Marko Grobelnik, Jan Odijk, Stelios Piperidis, Bente Maegaard, Joseph Mariani
PublisherEuropean Language Resources Association (ELRA)
Pages2240-2245
Number of pages6
ISBN (Electronic)9782951740891
Publication statusPublished - 2016
Externally publishedYes
Event10th International Conference on Language Resources and Evaluation, LREC 2016 - Portoroz, Slovenia
Duration: 23 May 201628 May 2016

Publication series

NameProceedings of the 10th International Conference on Language Resources and Evaluation, LREC 2016

Conference

Conference10th International Conference on Language Resources and Evaluation, LREC 2016
Country/TerritorySlovenia
CityPortoroz
Period23/05/1628/05/16

Keywords

  • Machine translation
  • Post-editing
  • Quality Estimation

Fingerprint

Dive into the research topics of 'Phrase-level segmentation and labelling of machine translation errors'. Together they form a unique fingerprint.

Cite this