Salvato in:
Dettagli Bibliografici
Autore principale: Fernando Martínez
Natura: Artículo científico
Lingua:en
Pubblicazione: Asociación Española para la Inteligencia Artificial 2004
Soggetti:
Accesso online:https://www.redalyc.org/articulo.oa?id=92582208
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866815730191695872
author Fernando Martínez
author_facet Fernando Martínez
contents SINAI experience at CLEF Fernando Martínez L. Alfonso Ureña M. Teresa Martín Ingeniería gies word Cross Multi trieval This paper reports our work on CLEF and Cross-Language Information Retrieval using CLEF re-sources. We aim to construct a highly language-independent CLIR model. To accomplish this ob-jective, several problems must be overcome: texttranslation or pseudo-translation and merging theobtained results for each language for a givenquery. Three issues of text-translation are inves-tigated: the impact of translation probabilities,automatic multi-word recognition, and the gen-eration of similarity thesauri from a Web corpus.Because the proposed model is query-translationdriven, it is necessary to merge several monolin-gual results in a unique multilingual list of docu-ments. To accomplish this task, we propose a newapproach, which we call 2-step RSV, and we showthat it performs better than more traditional ap-proaches 2004 artículo científico 1137-3601 https://www.redalyc.org/articulo.oa?id=92582208 en http://www.redalyc.org/revista.oa?id=925 Inteligencia Artificial. Revista Iberoamericana de Inteligencia Artificial application/pdf Asociación Española para la Inteligencia Artificial Inteligencia Artificial. Revista Iberoamericana de Inteligencia Artificial (España) Num.22 Vol.8
format Artículo científico
id redalyc_92582208
language en
publishDate 2004
publisher Asociación Española para la Inteligencia Artificial
spellingShingle SINAI experience at CLEF
Fernando Martínez
Ingeniería
gies
word
Cross
Multi
trieval
SINAI experience at CLEF Fernando Martínez L. Alfonso Ureña M. Teresa Martín Ingeniería gies word Cross Multi trieval This paper reports our work on CLEF and Cross-Language Information Retrieval using CLEF re-sources. We aim to construct a highly language-independent CLIR model. To accomplish this ob-jective, several problems must be overcome: texttranslation or pseudo-translation and merging theobtained results for each language for a givenquery. Three issues of text-translation are inves-tigated: the impact of translation probabilities,automatic multi-word recognition, and the gen-eration of similarity thesauri from a Web corpus.Because the proposed model is query-translationdriven, it is necessary to merge several monolin-gual results in a unique multilingual list of docu-ments. To accomplish this task, we propose a newapproach, which we call 2-step RSV, and we showthat it performs better than more traditional ap-proaches 2004 artículo científico 1137-3601 https://www.redalyc.org/articulo.oa?id=92582208 en http://www.redalyc.org/revista.oa?id=925 Inteligencia Artificial. Revista Iberoamericana de Inteligencia Artificial application/pdf Asociación Española para la Inteligencia Artificial Inteligencia Artificial. Revista Iberoamericana de Inteligencia Artificial (España) Num.22 Vol.8
title SINAI experience at CLEF
topic Ingeniería
gies
word
Cross
Multi
trieval
url https://www.redalyc.org/articulo.oa?id=92582208