Merging possibilistic networks

Authors:
Salem Benferhat
Affiliations:
CRIL-CNRS, Université d'Artois, Rue Jean Souvraz 62307 Lens Cedex, France. Email: benferhat@cril.univ-artois.fr
Venue:
Proceedings of the 2006 conference on ECAI 2006: 17th European Conference on Artificial Intelligence August 29 -- September 1, 2006, Riva del Garda, Italy
Year:
2006

Citing 3
Cited 0

Possibilistic logic

Handbook of logic in artificial intelligence and logic programming (vol. 3)
Integration of weighted knowledge bases

Artificial Intelligence
Introduction to Bayesian Networks

Introduction to Bayesian Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper deals with merging multiple-source uncertain pieces of information, which are encoded by means of possibilistic networks. We first show that the merging of possibilistic networks having the same graphical structure can be easily achieved in polynomial time. When possibilistic networks have different graphical structures we show that their fusion can also be efficiently done by extending initial possibilistic networks into a same common structure. We then address two important problems: how to deal with cycles, and how to solve the subnormalization problem which reflects conflicts between sources?