Using translation paraphrases from trilingual corpora to improve phrase-based statistical machine translation: A preliminary report

Francisco Guzmán Herrera, Leonardo Garrido Luna

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

Abstract

Statistical methods have proven to be very effective when addressing linguistic problems, specially when dealing with Machine Translation. Nevertheless, Statistical Machine Translation effectiveness is limited to situations where large amounts of training data are available. Therefore, the broader the coverage of a SMT system is, the better the chances to get a reasonable output are. In this paper we propose a method to improve quality of translations of a phrase-based Machine Translation system by extending phrase-tables with the use of translation paraphrases learned from a third language. Our experiments were done translating from Spanish to English pivoting through French.

Original languageEnglish
Title of host publicationProceedings - 2007 6th Mexican International Conference on Artificial Intelligence, Special Session, MICAI 2007
Pages163-172
Number of pages10
DOIs
Publication statusPublished - 2008
Externally publishedYes
Event2007 6th Mexican International Conference on Artificial Intelligence, Special Session, MICAI 2007 - Aguascalientes, Mexico
Duration: 4 Nov 200710 Nov 2007

Other

Other2007 6th Mexican International Conference on Artificial Intelligence, Special Session, MICAI 2007
CountryMexico
CityAguascalientes
Period4/11/0710/11/07

Fingerprint

Surface mount technology
Linguistics
Statistical methods
Experiments

ASJC Scopus subject areas

  • Artificial Intelligence

Cite this

Herrera, F. G., & Luna, L. G. (2008). Using translation paraphrases from trilingual corpora to improve phrase-based statistical machine translation: A preliminary report. In Proceedings - 2007 6th Mexican International Conference on Artificial Intelligence, Special Session, MICAI 2007 (pp. 163-172). [4659306] https://doi.org/10.1109/MICAI.2007.34

Using translation paraphrases from trilingual corpora to improve phrase-based statistical machine translation : A preliminary report. / Herrera, Francisco Guzmán; Luna, Leonardo Garrido.

Proceedings - 2007 6th Mexican International Conference on Artificial Intelligence, Special Session, MICAI 2007. 2008. p. 163-172 4659306.

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

Herrera, FG & Luna, LG 2008, Using translation paraphrases from trilingual corpora to improve phrase-based statistical machine translation: A preliminary report. in Proceedings - 2007 6th Mexican International Conference on Artificial Intelligence, Special Session, MICAI 2007., 4659306, pp. 163-172, 2007 6th Mexican International Conference on Artificial Intelligence, Special Session, MICAI 2007, Aguascalientes, Mexico, 4/11/07. https://doi.org/10.1109/MICAI.2007.34
Herrera FG, Luna LG. Using translation paraphrases from trilingual corpora to improve phrase-based statistical machine translation: A preliminary report. In Proceedings - 2007 6th Mexican International Conference on Artificial Intelligence, Special Session, MICAI 2007. 2008. p. 163-172. 4659306 https://doi.org/10.1109/MICAI.2007.34
Herrera, Francisco Guzmán ; Luna, Leonardo Garrido. / Using translation paraphrases from trilingual corpora to improve phrase-based statistical machine translation : A preliminary report. Proceedings - 2007 6th Mexican International Conference on Artificial Intelligence, Special Session, MICAI 2007. 2008. pp. 163-172
@inproceedings{4817c66dc5ad40d0b4c56323322c8a7e,
title = "Using translation paraphrases from trilingual corpora to improve phrase-based statistical machine translation: A preliminary report",
abstract = "Statistical methods have proven to be very effective when addressing linguistic problems, specially when dealing with Machine Translation. Nevertheless, Statistical Machine Translation effectiveness is limited to situations where large amounts of training data are available. Therefore, the broader the coverage of a SMT system is, the better the chances to get a reasonable output are. In this paper we propose a method to improve quality of translations of a phrase-based Machine Translation system by extending phrase-tables with the use of translation paraphrases learned from a third language. Our experiments were done translating from Spanish to English pivoting through French.",
author = "Herrera, {Francisco Guzm{\'a}n} and Luna, {Leonardo Garrido}",
year = "2008",
doi = "10.1109/MICAI.2007.34",
language = "English",
isbn = "9780769531243",
pages = "163--172",
booktitle = "Proceedings - 2007 6th Mexican International Conference on Artificial Intelligence, Special Session, MICAI 2007",

}

TY - GEN

T1 - Using translation paraphrases from trilingual corpora to improve phrase-based statistical machine translation

T2 - A preliminary report

AU - Herrera, Francisco Guzmán

AU - Luna, Leonardo Garrido

PY - 2008

Y1 - 2008

N2 - Statistical methods have proven to be very effective when addressing linguistic problems, specially when dealing with Machine Translation. Nevertheless, Statistical Machine Translation effectiveness is limited to situations where large amounts of training data are available. Therefore, the broader the coverage of a SMT system is, the better the chances to get a reasonable output are. In this paper we propose a method to improve quality of translations of a phrase-based Machine Translation system by extending phrase-tables with the use of translation paraphrases learned from a third language. Our experiments were done translating from Spanish to English pivoting through French.

AB - Statistical methods have proven to be very effective when addressing linguistic problems, specially when dealing with Machine Translation. Nevertheless, Statistical Machine Translation effectiveness is limited to situations where large amounts of training data are available. Therefore, the broader the coverage of a SMT system is, the better the chances to get a reasonable output are. In this paper we propose a method to improve quality of translations of a phrase-based Machine Translation system by extending phrase-tables with the use of translation paraphrases learned from a third language. Our experiments were done translating from Spanish to English pivoting through French.

UR - http://www.scopus.com/inward/record.url?scp=57749189806&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=57749189806&partnerID=8YFLogxK

U2 - 10.1109/MICAI.2007.34

DO - 10.1109/MICAI.2007.34

M3 - Conference contribution

AN - SCOPUS:57749189806

SN - 9780769531243

SP - 163

EP - 172

BT - Proceedings - 2007 6th Mexican International Conference on Artificial Intelligence, Special Session, MICAI 2007

ER -