ALEPH: an EBMT system based on the preservation of proportional analogies between sentences across languages

Yves Lepage; Etienne Denoual

ALEPH: an EBMT system based on the preservation of proportional analogies between sentences across languages

Yves Lepage, Etienne Denoual

Research output: Contribution to conference › Paper › peer-review

Abstract

We designed, implemented and assessed ALEPH, a pure example-based machine translation system. It strictly does not make any use of variables, templates or training, does not have any explicit transfer component, and does not require any preprocessing of the aligned examples. It relies on a specific operation: the resolution of analogical equations, that neutralizes translation divergences in an elegant way. Starting only from theoretical results, a system that is state-of-the-art with the top IWSLT 2004 results could be built in six month time. Evaluated on the Unrestricted Data track of IWSLT 2004, our system achieved second place in CE, and third place in JE (with best BLEU for this latter track). For this year’s evaluation campaign, the features of the system allowed its immediate application to all possible language pairs in the C-STAR tracks.

Original language	English
Pages	37-44
Number of pages	8
Publication status	Published - 2005
Externally published	Yes
Event	2nd International Workshop on Spoken Language Translation, IWSLT 2005 - Pittsburgh, United States Duration: 2005 Oct 24 → 2005 Oct 25

Conference

Conference	2nd International Workshop on Spoken Language Translation, IWSLT 2005
Country/Territory	United States
City	Pittsburgh
Period	05/10/24 → 05/10/25

ASJC Scopus subject areas

Language and Linguistics
Human-Computer Interaction
Linguistics and Language

Cite this

@conference{9f6023be8b074561bd51acfc7270dbcb,

title = "ALEPH: an EBMT system based on the preservation of proportional analogies between sentences across languages",

abstract = "We designed, implemented and assessed ALEPH, a pure example-based machine translation system. It strictly does not make any use of variables, templates or training, does not have any explicit transfer component, and does not require any preprocessing of the aligned examples. It relies on a specific operation: the resolution of analogical equations, that neutralizes translation divergences in an elegant way. Starting only from theoretical results, a system that is state-of-the-art with the top IWSLT 2004 results could be built in six month time. Evaluated on the Unrestricted Data track of IWSLT 2004, our system achieved second place in CE, and third place in JE (with best BLEU for this latter track). For this year{\textquoteright}s evaluation campaign, the features of the system allowed its immediate application to all possible language pairs in the C-STAR tracks.",

author = "Yves Lepage and Etienne Denoual",

note = "Funding Information: This research was supported in part by the National Institute of Information and Commnications Technology. Publisher Copyright: {\textcopyright} 2005 International Workshop on Spoken Language Translation, IWSLT 2005.; 2nd International Workshop on Spoken Language Translation, IWSLT 2005 ; Conference date: 24-10-2005 Through 25-10-2005",

year = "2005",

language = "English",

pages = "37--44",

}

TY - CONF

T1 - ALEPH

T2 - 2nd International Workshop on Spoken Language Translation, IWSLT 2005

AU - Lepage, Yves

AU - Denoual, Etienne

N1 - Funding Information: This research was supported in part by the National Institute of Information and Commnications Technology. Publisher Copyright: © 2005 International Workshop on Spoken Language Translation, IWSLT 2005.

PY - 2005

Y1 - 2005

N2 - We designed, implemented and assessed ALEPH, a pure example-based machine translation system. It strictly does not make any use of variables, templates or training, does not have any explicit transfer component, and does not require any preprocessing of the aligned examples. It relies on a specific operation: the resolution of analogical equations, that neutralizes translation divergences in an elegant way. Starting only from theoretical results, a system that is state-of-the-art with the top IWSLT 2004 results could be built in six month time. Evaluated on the Unrestricted Data track of IWSLT 2004, our system achieved second place in CE, and third place in JE (with best BLEU for this latter track). For this year’s evaluation campaign, the features of the system allowed its immediate application to all possible language pairs in the C-STAR tracks.

AB - We designed, implemented and assessed ALEPH, a pure example-based machine translation system. It strictly does not make any use of variables, templates or training, does not have any explicit transfer component, and does not require any preprocessing of the aligned examples. It relies on a specific operation: the resolution of analogical equations, that neutralizes translation divergences in an elegant way. Starting only from theoretical results, a system that is state-of-the-art with the top IWSLT 2004 results could be built in six month time. Evaluated on the Unrestricted Data track of IWSLT 2004, our system achieved second place in CE, and third place in JE (with best BLEU for this latter track). For this year’s evaluation campaign, the features of the system allowed its immediate application to all possible language pairs in the C-STAR tracks.

UR - http://www.scopus.com/inward/record.url?scp=85133342120&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85133342120&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:85133342120

SP - 37

EP - 44

Y2 - 24 October 2005 through 25 October 2005

ER -

ALEPH: an EBMT system based on the preservation of proportional analogies between sentences across languages

Abstract

Conference

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this