Accéder directement au contenu Accéder directement à la navigation
Chapitre d'ouvrage

Ontology Alignment Using Web Linked Ontologies as Background Knowledge

Abstract : This paper proposes an ontology matching method for aligning a source ontology with target ontologies already published and linked on the Linked Open Data (LOD) cloud. This method relies on the refinement of a set of input alignments generated by existing ontology matching methods. Since the ontologies to be aligned can be expressed in several representation languages with different levels of expressiveness and the existing ontology matching methods can only be applied to some representation languages, the first step of our method consists in applying existing matching methods to as many ontology variants as possible. We then propose to apply two main strategies to refine the initial alignment set: the removal of different kinds of ambiguities between correspondences and the use of the links published on the LOD. We illustrate our proposal in the field of life sciences and environment.
Type de document :
Chapitre d'ouvrage
Liste complète des métadonnées

Littérature citée [25 références]  Voir  Masquer  Télécharger

https://hal-agroparistech.archives-ouvertes.fr/hal-01508810
Déposant : Juliette Dibie <>
Soumis le : vendredi 15 juin 2018 - 12:17:51
Dernière modification le : mercredi 18 novembre 2020 - 19:02:02
Archivage à long terme le : : dimanche 16 septembre 2018 - 13:23:36

Fichier

AKDM_6_paper 2_Hecht14.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

Citation

Thomas Hecht, Patrice Buche, Juliette Dibie-Barthelemy, Liliana Ibanescu, Cássia Trojahn dos Santos. Ontology Alignment Using Web Linked Ontologies as Background Knowledge. Fabrice Guillet; Bruno Pinaud; Gilles Venturini. Advances in Knowledge Discovery and Management, 665, Springer, pp.207-227, 2017, Studies in Computational Intelligence, 978-3-319-45762-8. ⟨10.1007/978-3-319-45763-5_11⟩. ⟨hal-01508810⟩

Partager

Métriques

Consultations de la notice

1589

Téléchargements de fichiers

983