Abstract
Sub-sentential alignment is the process by which multi-word translation units are extracted from sentence-aligned multilingual parallel texts. This process is required, for instance, in the course of training statistical machine translation systems. Standard approaches typically rely on the estimation of several probabilistic models of increasing complexity and on the use of various heuristics, that make it possible to align, first isolated words, then, by extension, groups of words. In this paper, we explore an alternative approach which relies on a much simpler principle: the comparison of occurrence profiles in sub-corpora obtained by sampling. After analyzing the strengths and weaknesses of this approach, we show how to improve the detection of multi-word translation units and evaluate these improvements on machine translation tasks.
| Original language | English |
|---|---|
| Pages (from-to) | 1-23 |
| Number of pages | 23 |
| Journal | Machine Translation |
| Volume | 27 |
| Issue number | 1 |
| DOIs | |
| Publication status | Published - 2013 Mar |
Fingerprint
Keywords
- Association measures
- Phrase-based machine translation
- Sub-sentential alignment
ASJC Scopus subject areas
- Artificial Intelligence
- Software
- Language and Linguistics
- Linguistics and Language
Cite this
Generalizing sampling-based multilingual alignment. / Lardilleux, Adrien; Yvon, François; Lepage, Yves.
In: Machine Translation, Vol. 27, No. 1, 03.2013, p. 1-23.Research output: Contribution to journal › Article
}
TY - JOUR
T1 - Generalizing sampling-based multilingual alignment
AU - Lardilleux, Adrien
AU - Yvon, François
AU - Lepage, Yves
PY - 2013/3
Y1 - 2013/3
N2 - Sub-sentential alignment is the process by which multi-word translation units are extracted from sentence-aligned multilingual parallel texts. This process is required, for instance, in the course of training statistical machine translation systems. Standard approaches typically rely on the estimation of several probabilistic models of increasing complexity and on the use of various heuristics, that make it possible to align, first isolated words, then, by extension, groups of words. In this paper, we explore an alternative approach which relies on a much simpler principle: the comparison of occurrence profiles in sub-corpora obtained by sampling. After analyzing the strengths and weaknesses of this approach, we show how to improve the detection of multi-word translation units and evaluate these improvements on machine translation tasks.
AB - Sub-sentential alignment is the process by which multi-word translation units are extracted from sentence-aligned multilingual parallel texts. This process is required, for instance, in the course of training statistical machine translation systems. Standard approaches typically rely on the estimation of several probabilistic models of increasing complexity and on the use of various heuristics, that make it possible to align, first isolated words, then, by extension, groups of words. In this paper, we explore an alternative approach which relies on a much simpler principle: the comparison of occurrence profiles in sub-corpora obtained by sampling. After analyzing the strengths and weaknesses of this approach, we show how to improve the detection of multi-word translation units and evaluate these improvements on machine translation tasks.
KW - Association measures
KW - Phrase-based machine translation
KW - Sub-sentential alignment
UR - http://www.scopus.com/inward/record.url?scp=84875743779&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84875743779&partnerID=8YFLogxK
U2 - 10.1007/s10590-012-9126-0
DO - 10.1007/s10590-012-9126-0
M3 - Article
AN - SCOPUS:84875743779
VL - 27
SP - 1
EP - 23
JO - Machine Translation
JF - Machine Translation
SN - 0922-6567
IS - 1
ER -