The GREYC/LLACAN Machine Translation Systems for the IWSLT 2010 Campaign

Julien Gosme; Wigdan Mekki; Fathi Debili; Yves Lepage; Nadine Lucas

The GREYC/LLACAN Machine Translation Systems for the IWSLT 2010 Campaign

Julien Gosme, Wigdan Mekki, Fathi Debili, Yves Lepage, Nadine Lucas

研究成果: Paper › 査読

抄録

In this paper we explore the contribution of the use of two Arabic morphological analyzers as preprocessing tools for statistical machine translation. Similar investigations have already been reported for morphologically rich languages like German, Turkish and Arabic. Here, we focus on the case of the Arabic language and mainly discuss the use of the G-LexAr analyzer. A preliminary experiment has been designed to choose the most promising translation system among the 3 G-LexAr-based systems, we concluded that the systems are equivalent. Nevertheless, we decided to use the lemmatized output of G-LexAr and use its translations as primary run for the BTEC AE track. The results showed that G-LexAr outputs degrades translation compared to the basic SMT system trained on the un-analyzed corpus.

本文言語	English
ページ	59-65
ページ数	7
出版ステータス	Published - 2010
外部発表	はい
イベント	7th International Workshop on Spoken Language Translation, IWSLT 2010 - Paris, France 継続期間: 2010 12月 2 → 2010 12月 3

Conference

Conference	7th International Workshop on Spoken Language Translation, IWSLT 2010
国/地域	France
City	Paris
Period	10/12/2 → 10/12/3

ASJC Scopus subject areas

言語および言語学
人間とコンピュータの相互作用
言語学および言語

他のファイルとリンク

引用スタイル

@conference{d97fe95bb0064f1f8e7957f1365f3ebc,

title = "The GREYC/LLACAN Machine Translation Systems for the IWSLT 2010 Campaign",

abstract = "In this paper we explore the contribution of the use of two Arabic morphological analyzers as preprocessing tools for statistical machine translation. Similar investigations have already been reported for morphologically rich languages like German, Turkish and Arabic. Here, we focus on the case of the Arabic language and mainly discuss the use of the G-LexAr analyzer. A preliminary experiment has been designed to choose the most promising translation system among the 3 G-LexAr-based systems, we concluded that the systems are equivalent. Nevertheless, we decided to use the lemmatized output of G-LexAr and use its translations as primary run for the BTEC AE track. The results showed that G-LexAr outputs degrades translation compared to the basic SMT system trained on the un-analyzed corpus.",

author = "Julien Gosme and Wigdan Mekki and Fathi Debili and Yves Lepage and Nadine Lucas",

year = "2010",

language = "English",

pages = "59--65",

}

TY - CONF

T1 - The GREYC/LLACAN Machine Translation Systems for the IWSLT 2010 Campaign

AU - Gosme, Julien

AU - Mekki, Wigdan

AU - Debili, Fathi

AU - Lepage, Yves

AU - Lucas, Nadine

PY - 2010

Y1 - 2010

N2 - In this paper we explore the contribution of the use of two Arabic morphological analyzers as preprocessing tools for statistical machine translation. Similar investigations have already been reported for morphologically rich languages like German, Turkish and Arabic. Here, we focus on the case of the Arabic language and mainly discuss the use of the G-LexAr analyzer. A preliminary experiment has been designed to choose the most promising translation system among the 3 G-LexAr-based systems, we concluded that the systems are equivalent. Nevertheless, we decided to use the lemmatized output of G-LexAr and use its translations as primary run for the BTEC AE track. The results showed that G-LexAr outputs degrades translation compared to the basic SMT system trained on the un-analyzed corpus.

AB - In this paper we explore the contribution of the use of two Arabic morphological analyzers as preprocessing tools for statistical machine translation. Similar investigations have already been reported for morphologically rich languages like German, Turkish and Arabic. Here, we focus on the case of the Arabic language and mainly discuss the use of the G-LexAr analyzer. A preliminary experiment has been designed to choose the most promising translation system among the 3 G-LexAr-based systems, we concluded that the systems are equivalent. Nevertheless, we decided to use the lemmatized output of G-LexAr and use its translations as primary run for the BTEC AE track. The results showed that G-LexAr outputs degrades translation compared to the basic SMT system trained on the un-analyzed corpus.

UR - http://www.scopus.com/inward/record.url?scp=85133186204&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85133186204&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:85133186204

SP - 59

EP - 65

T2 - 7th International Workshop on Spoken Language Translation, IWSLT 2010

Y2 - 2 December 2010 through 3 December 2010

ER -

The GREYC/LLACAN Machine Translation Systems for the IWSLT 2010 Campaign

抄録

Conference

ASJC Scopus subject areas

他のファイルとリンク

フィンガープリント

引用スタイル