Alignment symmetrization optimization targeting phrase pivot statistical machine translation

Ahmed El Kholy, Nizar Habash

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

An important step in mainstream statistical machine translation (SMT) is combining bidirectional alignments into one alignment model. This process is called symmetrization. Most of the symmetrization heuristics and models are focused on direct translation (source-to-target). In this paper, we present symmetrization heuristic relaxation to improve the quality of phrase-pivot SMT (source-[pivot]-target). We show positive results (1.2 BLEU points) on Hebrew-to-Arabic SMT pivoting on English.

Original languageEnglish (US)
Title of host publicationProceedings of the 17th Annual Conference of the European Association for Machine Translation, EAMT 2014
PublisherEuropean Association for Machine Translation
Pages63-70
Number of pages8
ISBN (Electronic)9789535537533
StatePublished - Jan 1 2014
Event17th Annual Conference of the European Association for Machine Translation, EAMT 2014 - Dubrovnik, Croatia
Duration: Jun 16 2014Jun 18 2014

Other

Other17th Annual Conference of the European Association for Machine Translation, EAMT 2014
CountryCroatia
CityDubrovnik
Period6/16/146/18/14

Fingerprint

Statistical Machine Translation
Alignment
Heuristics

ASJC Scopus subject areas

  • Language and Linguistics
  • Human-Computer Interaction
  • Software

Cite this

El Kholy, A., & Habash, N. (2014). Alignment symmetrization optimization targeting phrase pivot statistical machine translation. In Proceedings of the 17th Annual Conference of the European Association for Machine Translation, EAMT 2014 (pp. 63-70). European Association for Machine Translation.

Alignment symmetrization optimization targeting phrase pivot statistical machine translation. / El Kholy, Ahmed; Habash, Nizar.

Proceedings of the 17th Annual Conference of the European Association for Machine Translation, EAMT 2014. European Association for Machine Translation, 2014. p. 63-70.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

El Kholy, A & Habash, N 2014, Alignment symmetrization optimization targeting phrase pivot statistical machine translation. in Proceedings of the 17th Annual Conference of the European Association for Machine Translation, EAMT 2014. European Association for Machine Translation, pp. 63-70, 17th Annual Conference of the European Association for Machine Translation, EAMT 2014, Dubrovnik, Croatia, 6/16/14.
El Kholy A, Habash N. Alignment symmetrization optimization targeting phrase pivot statistical machine translation. In Proceedings of the 17th Annual Conference of the European Association for Machine Translation, EAMT 2014. European Association for Machine Translation. 2014. p. 63-70
El Kholy, Ahmed ; Habash, Nizar. / Alignment symmetrization optimization targeting phrase pivot statistical machine translation. Proceedings of the 17th Annual Conference of the European Association for Machine Translation, EAMT 2014. European Association for Machine Translation, 2014. pp. 63-70
@inproceedings{7af56611d00a415ea1afd497dd2667ec,
title = "Alignment symmetrization optimization targeting phrase pivot statistical machine translation",
abstract = "An important step in mainstream statistical machine translation (SMT) is combining bidirectional alignments into one alignment model. This process is called symmetrization. Most of the symmetrization heuristics and models are focused on direct translation (source-to-target). In this paper, we present symmetrization heuristic relaxation to improve the quality of phrase-pivot SMT (source-[pivot]-target). We show positive results (1.2 BLEU points) on Hebrew-to-Arabic SMT pivoting on English.",
author = "{El Kholy}, Ahmed and Nizar Habash",
year = "2014",
month = "1",
day = "1",
language = "English (US)",
pages = "63--70",
booktitle = "Proceedings of the 17th Annual Conference of the European Association for Machine Translation, EAMT 2014",
publisher = "European Association for Machine Translation",

}

TY - GEN

T1 - Alignment symmetrization optimization targeting phrase pivot statistical machine translation

AU - El Kholy, Ahmed

AU - Habash, Nizar

PY - 2014/1/1

Y1 - 2014/1/1

N2 - An important step in mainstream statistical machine translation (SMT) is combining bidirectional alignments into one alignment model. This process is called symmetrization. Most of the symmetrization heuristics and models are focused on direct translation (source-to-target). In this paper, we present symmetrization heuristic relaxation to improve the quality of phrase-pivot SMT (source-[pivot]-target). We show positive results (1.2 BLEU points) on Hebrew-to-Arabic SMT pivoting on English.

AB - An important step in mainstream statistical machine translation (SMT) is combining bidirectional alignments into one alignment model. This process is called symmetrization. Most of the symmetrization heuristics and models are focused on direct translation (source-to-target). In this paper, we present symmetrization heuristic relaxation to improve the quality of phrase-pivot SMT (source-[pivot]-target). We show positive results (1.2 BLEU points) on Hebrew-to-Arabic SMT pivoting on English.

UR - http://www.scopus.com/inward/record.url?scp=85000538628&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85000538628&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:85000538628

SP - 63

EP - 70

BT - Proceedings of the 17th Annual Conference of the European Association for Machine Translation, EAMT 2014

PB - European Association for Machine Translation

ER -