How can ontologies give you clue for truth-discovery? an exploratory study

Abstract : The main aim of truth-finding methods is to identify the most reliable and trustworthy data among a set of facts. Since existing methods assume a single true value, they cannot deal with numerous real-world use cases in which a set of true values exists for a given fact, even for functional predicate (e.g. Picasso is born in Màlaga and in Spain). This paper studies how traditional truth-finding methods can be adapted to this setting. After introducing a new definition of true value and discussing associated implications, we propose an approach that can be used to identify true values among a set of non-conflicting claims; it takes advantage of belief functions to incorporate knowledge about value relationships in the form of a partial ordering of claimed values. By reducing the error rate up to 30% adapting classical approaches, the effectiveness and suitability of our proposal is clearly highlighted through empirical experiments performed on DBpedia.
Type de document :
Communication dans un congrès
WIMS '16 Proceedings of the 6th International Conference on Web Intelligence, Mining and Semantics, Jun 2016, Nîmes, France. 〈10.1145/2912845.2912848〉
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-01484980
Contributeur : Sébastien Harispe <>
Soumis le : mercredi 8 mars 2017 - 09:37:12
Dernière modification le : jeudi 15 novembre 2018 - 10:42:04

Identifiants

Citation

Valentina Beretta, Sébastien Harispe, Sylvie Ranwez, Isabelle Mougenot. How can ontologies give you clue for truth-discovery? an exploratory study. WIMS '16 Proceedings of the 6th International Conference on Web Intelligence, Mining and Semantics, Jun 2016, Nîmes, France. 〈10.1145/2912845.2912848〉. 〈hal-01484980〉

Partager

Métriques

Consultations de la notice

202