Normal view MARC view ISBD view

Building and using comparable corpora for domain-specific bilingual lexicon extraction / Fišer, Darja ; Ljubešić, Nikola ; Vintar, Špela ; Pollak, Senja.

By: Fišer, Darja.
Contributor(s): Vintar, Špela [aut] | Pollak, Senja [aut] | Ljubešić, Nikola, informatičar [aut].
Material type: ArticleArticleDescription: 19-26 str.Other title: Building and using comparable corpora for domain-specific bilingual lexicon extraction [Naslov na engleskom:].Subject(s): 5.04 | comparable corpora, bilingual lexicon extraction, domain lexicons hrv | comparable corpora, bilingual lexicon extraction, domain lexicons engOnline resources: Elektronička verzija In: 4th Workshop on Building and Using Comparable Corpora: Comparable Corpora and the Web (24.7.2011. ; Portland, SAD) 4th Workshop on Building and Using Comparable Corpora: Comparable Corpora and the Web str. 19-26Summary: This paper presents a series of experiments aimed at inducing and evaluating domain- specific bilingual lexica from comparable corpora. First, a small English-Slovene comparable corpus from health magazines was manually constructed and then used to compile a large comparable corpus on health-related topics from web corpora. Next, a bilingual lexicon for the domain was extracted from the corpus by comparing context vectors in the two languages. Evaluation of the results shows that a 2-way translation of context vectors significantly improves precision of the extracted translation equivalents. We also show that it is sufficient to increase the corpus for one language in order to obtain a higher recall, and that the increase of the number of new words is linear in the size of the corpus. Finally, we demonstrate that by lowering the frequency threshold for context vectors, the drop in precision is much slower than the increase of recall.
Tags from this library: No tags from this library for this title. Log in to add tags.
No physical items for this record

This paper presents a series of experiments aimed at inducing and evaluating domain- specific bilingual lexica from comparable corpora. First, a small English-Slovene comparable corpus from health magazines was manually constructed and then used to compile a large comparable corpus on health-related topics from web corpora. Next, a bilingual lexicon for the domain was extracted from the corpus by comparing context vectors in the two languages. Evaluation of the results shows that a 2-way translation of context vectors significantly improves precision of the extracted translation equivalents. We also show that it is sufficient to increase the corpus for one language in order to obtain a higher recall, and that the increase of the number of new words is linear in the size of the corpus. Finally, we demonstrate that by lowering the frequency threshold for context vectors, the drop in precision is much slower than the increase of recall.

Projekt MZOS 130-1301679-1380

ENG

There are no comments for this item.

Log in to your account to post a comment.

Powered by Koha

//