Skip to Main content Skip to Navigation
Conference papers

OBIRS-feedback, une méthode de reformulation utilisant une ontologie de domaine

Abstract : The lack of accuracy of an information retrieval system (IRS) may be due to an inadequate formulation of user’s queries. Reformulation or query expansion is a possible solution to this problem. In this paper, we introduce a reformulation method based upon a domain ontology. This conceptual relevance feedback method uses a set of documents a user has deemed relevant to search a set of concepts that maximizes IRS performances. Those performances are assessed in an original way, using indicators that are formalized. This method has been evaluated using our environment OBIRS (Ontology Based Information Retrieval System) as base system, MeSH as domain ontology and MuCHMORE as a test collection.
Complete list of metadata

https://hal.inrae.fr/hal-02746138
Contributor : Migration Prodinra <>
Submitted on : Wednesday, June 3, 2020 - 9:34:59 AM
Last modification on : Wednesday, September 8, 2021 - 3:53:27 AM

Identifiers

  • HAL Id : hal-02746138, version 1
  • PRODINRA : 49627

Citation

Mohameth François Sy, Sylvie Ranwez, Vincent Ranwez, Jacky Montmain. OBIRS-feedback, une méthode de reformulation utilisant une ontologie de domaine. Actes de la neuvième édition de la COnférence en Recherche d'Information et Applications CORIA 2012, Mar 2012, Bordeaux, France. ⟨hal-02746138⟩

Share

Metrics

Record views

38