DI-UMONS : Dépôt institutionnel de l’université de Mons

Recherche transversale
Rechercher
(titres de publication, de périodique et noms de colloque inclus)
2002-02-01 - Article/Dans un journal avec peer-review - Anglais - 19 page(s)

Fortemps Philippe , Slowinski R., "A graded quadrivalent logic for preference modelling: Loyola-like approach" in Fuzzy Optimization & Decision Making, 1, 1, 93-111

  • Edition : Springer Science & Business Media B.V.
  • Codes CREF : Intelligence artificielle (DI1180), Modèles mathématiques d'aide à la décision (DI1151), Logiques non classiques (DI1171)
  • Unités de recherche UMONS : Mathématique et Recherche opérationnelle (F151)
  • Instituts UMONS : Institut de Recherche en Technologies de l’Information et Sciences de l’Informatique (InforTech)
Texte intégral :

Abstract(s) :

(Anglais) We extend a quadrivalent logic of Belnap to graded truth values in order to handle graded relevance of positive and negative arguments provided in preferential information concerning ranking of a finite set of alternatives. This logic is used to design the preference modelling and exploitation phases of decision aiding with respect to the ranking problem. The graded arguments are presented on an ordinal scale and their aggregation leads to preference model in form of four graded outranking relations (true, false, unknown and contradictory). The exploitation procedure combines the min-scoring procedure with the leximin rule. Aggregation of positive and negative arguments as well as exploitation of the resulting outranking relations is concordant with an advice given by St. Ignatius of Loyola (1548) how to make a good choice.

Identifiants :
  • DOI : 10.1023/A:1013731910441