Proceedings of the 13th Workshop on Multiword Expressions (MWE 2017) 2017
DOI: 10.18653/v1/w17-1711
|View full text |Cite
|
Sign up to set email alerts
|

Discovering Light Verb Constructions and their Translations from Parallel Corpora without Word Alignment

Abstract: We propose a method for joint unsupervised discovery of multiword expressions (MWEs) and their translations from parallel corpora. First, we apply independent monolingual MWE extraction in source and target languages simultaneously. Then, we calculate translation probability, association score and distributional similarity of co-occurring pairs. Finally, we rank all translations of a given MWE using a linear combination of these features. Preliminary experiments on light verb constructions show promising resul… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1

Citation Types

0
1
0

Year Published

2019
2019
2020
2020

Publication Types

Select...
3

Relationship

1
2

Authors

Journals

citations
Cited by 3 publications
(1 citation statement)
references
References 12 publications
0
1
0
Order By: Relevance
“…For example, one can use automatic word alignment tools to locate phrases in which the alignment is not perfect, indicating the presence of a phraseological unit (de Medeiros Caseli et al, 2009;Tsvetkov and Wintner, 2011;Salehi and Cook, 2013). Additionally, it is possible to use a method similar to the one for variant generation for translation generation: if a candidate cannot be translated word-by-word, then it is probably a phraseological unit (Morin et al, 2007;Vargas et al, 2017). Translation-based scores are not available in the mwetoolkit, but this is intended as future work.…”
Section: Other Scoresmentioning
confidence: 99%
“…For example, one can use automatic word alignment tools to locate phrases in which the alignment is not perfect, indicating the presence of a phraseological unit (de Medeiros Caseli et al, 2009;Tsvetkov and Wintner, 2011;Salehi and Cook, 2013). Additionally, it is possible to use a method similar to the one for variant generation for translation generation: if a candidate cannot be translated word-by-word, then it is probably a phraseological unit (Morin et al, 2007;Vargas et al, 2017). Translation-based scores are not available in the mwetoolkit, but this is intended as future work.…”
Section: Other Scoresmentioning
confidence: 99%