Fuzzy Syntactic Reordering for Phrase-based Statistical Machine Translation

Jacob Andreas; Nizar Habash; Owen Rambow

Fuzzy Syntactic Reordering for Phrase-based Statistical Machine Translation

Jacob Andreas, Nizar Habash, Owen Rambow

Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

The quality of Arabic-English statistical machine translation often suffers as a result of standard phrase-based SMT systems' inability to perform long-range re-orderings, specifically those needed to translate VSO-ordered Arabic sentences. This problem is further exacerbated by the low performance of Arabic parsers on subject and subject span detection. In this paper, we present two parse “fuzzification” techniques which allow the translation system to select among a range of possible S-V re-orderings. With this approach, we demonstrate a 0.3-point improvement in BLEU score (69% of the maximum possible using gold parses), and a corresponding improvement in the percentage of syntactically well-formed subjects under a manual evaluation.

Original language	English (US)
Title of host publication	WMT 2011 - 6thWorkshop on Statistical Machine Translation, Proceedings of the Workshop
Publisher	Association for Computational Linguistics (ACL)
Pages	227-236
Number of pages	10
ISBN (Electronic)	9781937284121
State	Published - 2011
Event	6thWorkshop on Statistical Machine Translation, WMT 2011 - Edinburgh, United Kingdom Duration: Jul 30 2011 → Jul 31 2011

Publication series

Name	WMT 2011 - 6thWorkshop on Statistical Machine Translation, Proceedings of the Workshop

Conference

Conference	6thWorkshop on Statistical Machine Translation, WMT 2011
Country/Territory	United Kingdom
City	Edinburgh
Period	7/30/11 → 7/31/11

ASJC Scopus subject areas

Human-Computer Interaction
Language and Linguistics
Software

Cite this

Fuzzy Syntactic Reordering for Phrase-based Statistical Machine Translation. / Andreas, Jacob; Habash, Nizar; Rambow, Owen.
WMT 2011 - 6thWorkshop on Statistical Machine Translation, Proceedings of the Workshop. Association for Computational Linguistics (ACL), 2011. p. 227-236 (WMT 2011 - 6thWorkshop on Statistical Machine Translation, Proceedings of the Workshop).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Andreas, J, Habash, N & Rambow, O 2011, Fuzzy Syntactic Reordering for Phrase-based Statistical Machine Translation. in WMT 2011 - 6thWorkshop on Statistical Machine Translation, Proceedings of the Workshop. WMT 2011 - 6thWorkshop on Statistical Machine Translation, Proceedings of the Workshop, Association for Computational Linguistics (ACL), pp. 227-236, 6thWorkshop on Statistical Machine Translation, WMT 2011, Edinburgh, United Kingdom, 7/30/11.

Andreas, Jacob ; Habash, Nizar ; Rambow, Owen. / Fuzzy Syntactic Reordering for Phrase-based Statistical Machine Translation. WMT 2011 - 6thWorkshop on Statistical Machine Translation, Proceedings of the Workshop. Association for Computational Linguistics (ACL), 2011. pp. 227-236 (WMT 2011 - 6thWorkshop on Statistical Machine Translation, Proceedings of the Workshop).

@inproceedings{0e08792dddfe407aae4837504b7d7d54,

title = "Fuzzy Syntactic Reordering for Phrase-based Statistical Machine Translation",

abstract = "The quality of Arabic-English statistical machine translation often suffers as a result of standard phrase-based SMT systems' inability to perform long-range re-orderings, specifically those needed to translate VSO-ordered Arabic sentences. This problem is further exacerbated by the low performance of Arabic parsers on subject and subject span detection. In this paper, we present two parse “fuzzification” techniques which allow the translation system to select among a range of possible S-V re-orderings. With this approach, we demonstrate a 0.3-point improvement in BLEU score (69% of the maximum possible using gold parses), and a corresponding improvement in the percentage of syntactically well-formed subjects under a manual evaluation.",

author = "Jacob Andreas and Nizar Habash and Owen Rambow",

note = "Publisher Copyright: {\textcopyright} 2011 Association for Computational Linguistics; 6thWorkshop on Statistical Machine Translation, WMT 2011 ; Conference date: 30-07-2011 Through 31-07-2011",

year = "2011",

language = "English (US)",

series = "WMT 2011 - 6thWorkshop on Statistical Machine Translation, Proceedings of the Workshop",

publisher = "Association for Computational Linguistics (ACL)",

pages = "227--236",

booktitle = "WMT 2011 - 6thWorkshop on Statistical Machine Translation, Proceedings of the Workshop",

}

TY - GEN

T1 - Fuzzy Syntactic Reordering for Phrase-based Statistical Machine Translation

AU - Andreas, Jacob

AU - Habash, Nizar

AU - Rambow, Owen

PY - 2011

Y1 - 2011

N2 - The quality of Arabic-English statistical machine translation often suffers as a result of standard phrase-based SMT systems' inability to perform long-range re-orderings, specifically those needed to translate VSO-ordered Arabic sentences. This problem is further exacerbated by the low performance of Arabic parsers on subject and subject span detection. In this paper, we present two parse “fuzzification” techniques which allow the translation system to select among a range of possible S-V re-orderings. With this approach, we demonstrate a 0.3-point improvement in BLEU score (69% of the maximum possible using gold parses), and a corresponding improvement in the percentage of syntactically well-formed subjects under a manual evaluation.

AB - The quality of Arabic-English statistical machine translation often suffers as a result of standard phrase-based SMT systems' inability to perform long-range re-orderings, specifically those needed to translate VSO-ordered Arabic sentences. This problem is further exacerbated by the low performance of Arabic parsers on subject and subject span detection. In this paper, we present two parse “fuzzification” techniques which allow the translation system to select among a range of possible S-V re-orderings. With this approach, we demonstrate a 0.3-point improvement in BLEU score (69% of the maximum possible using gold parses), and a corresponding improvement in the percentage of syntactically well-formed subjects under a manual evaluation.

UR - http://www.scopus.com/inward/record.url?scp=84878177035&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84878177035&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84878177035

T3 - WMT 2011 - 6thWorkshop on Statistical Machine Translation, Proceedings of the Workshop

SP - 227

EP - 236

BT - WMT 2011 - 6thWorkshop on Statistical Machine Translation, Proceedings of the Workshop

PB - Association for Computational Linguistics (ACL)

T2 - 6thWorkshop on Statistical Machine Translation, WMT 2011

Y2 - 30 July 2011 through 31 July 2011

ER -

Fuzzy Syntactic Reordering for Phrase-based Statistical Machine Translation

Abstract

Publication series

Conference

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this