Please use this identifier to cite or link to this item: http://hdl.handle.net/10609/151509
Title: IT: Moving towards real multilingualism
Author: Oliver, Antoni  
Borrell Condom, Cristina
Citation: Oliver, A. [Antoni] & Borrell, C. [Cristina]. (2009). IT: Moving Towards Real Multilingualism. NLP, Corpus Linguistics, Corpus Based Grammar Research. Fifth International Conference Smolenice, Eslovàquia 25-27 novembre de 2009 Proceedings (p. 279-286). Bratislava: Tribun
Abstract: The Linguamón-UOC Chair in Multilingualism of the Universitat Oberta de Catalunya has developed a project consisting of the automatic elaboration of linguistic resources, those including Catalan. For this, we have created an automatic extractor of terminology, which is freely distributed, multi-platform and adaptable to users' needs. One of its most important useful applications is the elaboration of glossaries, both monolingual and multilingual, based on a set of documents. The system automatically extracts the term candidates from the texts introduced by the user. Later, a specialist reviews the list of lexical units to verify the result. In the Chair in Multilingualism we have applied this tool to the expansion of the Eurovoc glossary with its Catalan version. We have extracted the equivalents for this language from the entries in Spanish and a bilingual Spanish – Catalan parallel corpus. From the application of the extractor, we obtained a multilingual glossary of 2 531 terms involving EU policies. In a second stage, we have used different techniques. Firstly, starting from the terms that were not yet translated by the first step, we have generated hypothesis of the Catalan translation of the Spanish Eurovoc terms, thus by the compositionality method. Secondly, we have applied statistically automatic translation, and finally we have checked the results in a monolingual corpus or by means of a restricted Web search. Our aim is to translate automatically as much terms as possible. With this project we try to demonstrate that having the suitable resources can remarkably reduce the translation costs.
Document type: info:eu-repo/semantics/conferenceObject
Issue Date: Nov-2009
Publication license: http://creativecommons.org/licenses/by-nc-nd/3.0/es/  
Appears in Collections:Conferencias

Files in This Item:
File Description SizeFormat 
2007-MovingTowards-Oliver-Borrell.pdf61,82 kBAdobe PDFThumbnail
View/Open
Share:
Export:
View statistics

This item is licensed under aCreative Commons License Creative Commons