Event-based textual document retrieval by using semantic role labeling and coreference resolution

Chia-Hung Lin 1 Chia-Wei Yen 1 Jen-Shin Hong 1 Samuel Cruz-Lara 2
1 National ChiNan University
2 TALARIS - Natural Language Processing: representation, inference and semantics
Inria Nancy - Grand Est, LORIA - Laboratoire Lorrain de Recherche en Informatique et ses Applications
Abstract : Conventional keyword-based indexing and retrieval techniques for textual documents lack of precision when a long query string is employed in order to discover documents containing a specific “event”, such as “Einstein discovered relativity”. This paper proposes a framework to resolve such a problem. In our proposal, we apply semantic role labeling and coreference techniques in order to parse each sentence within textual documents into three elements: subject, object and predicates. These elements can subsequently be used for indexing and retrieval. Our primitive evaluation experiments have shown that this promising methodology raises the retrieval precision if we compared it to conventional literal termmatching techniques.
Keywords : event-based textual document retrieval semantic role labeling coreference resolution
Type de document :
Communication dans un congrès
IADIS International Conference WWW/Internet 2007, Oct 2007, Vila Real, Portugal. 2007
Liste complète des métadonnées
https://hal.inria.fr/inria-00192465
Contributeur : Samuel Cruz-Lara <>
Soumis le : mercredi 28 novembre 2007 - 10:45:43
Dernière modification le : jeudi 11 janvier 2018 - 06:21:35
Document(s) archivé(s) le : lundi 12 avril 2010 - 05:21:54

Fichier

IADIS-event-search-final.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : inria-00192465, version 1

Collections

INRIA | LORIA | UNIV-LORRAINE

Citation

Chia-Hung Lin, Chia-Wei Yen, Jen-Shin Hong, Samuel Cruz-Lara. Event-based textual document retrieval by using semantic role labeling and coreference resolution. IADIS International Conference WWW/Internet 2007, Oct 2007, Vila Real, Portugal. 2007. 〈inria-00192465〉

Exporter

BibTeX TEI DC DCterms EndNote

Partager

Métriques

Consultations de la notice

210

Téléchargements de fichiers

185

Contact


archive-ouverte@inria.fr