Logical representation and fusion of prioritized information based on guaranteed possibility measures: application to the distance-based merging of classical bases

  • Authors:

  • Salem Benferhat;Souhila Kaci

  • Affiliations:

  • Centre de Recherche en Informatique de Lens (C.R.I.L.)- C.N.R.S., Université d'Artois, Rue Jean Souvraz, SP 18, 62307 Lens Cedex, France;Institut de Recherche en Informatique de Toulouse (I.R.I.T.)- C.N.R.S., Université Paul Sabatier, 118 route de Narbonne, 31062 Toulouse Cedex 4, France

  • Venue:
  • Artificial Intelligence - Special issue: Fuzzy set and possibility theory-based methods in artificial intelligence
  • Year:
  • 2003

Quantified Score

Hi-index 0.00

Visualization

Abstract

In the possibility theory framework, prioritized information can be logically expressed in different formats. The most usual way, used in standard possibilistic logic, is to associate necessity degrees with propositional formulas. This paper considers another representation and fusion of priorirized information using guaranteed possibility measures. Prioritized pieces of information are then represented by sets of weighted formulas, called Δ-knowledge bases, where weights are lower bounds of guaranteed possibility degrees of formulas.We first show that the basic notions of standard possibilistic logic have natural counterparts when dealing with Δ-knowledge bases. In particular we present the inference machinery, and provide syntactic, but semantically meaningful, merging of Δ-knowledge bases. In the second part of the paper, we show that distance-based merging propositional knowledge bases can be naturally encoded using Δ-knowledge bases. Moreover, this encoding is more efficient than the necessity-based encoding of distance-based merging operator.