Dempster Specialization Matrices and the Combination of Belief Functions

Authors:
Paul-André Monney
Affiliations:
-
Venue:
ECSQARU '01 Proceedings of the 6th European Conference on Symbolic and Quantitative Approaches to Reasoning with Uncertainty
Year:
2001

Citing 3
Cited 0

The Combination of Evidence in the Transferable Belief Model

IEEE Transactions on Pattern Analysis and Machine Intelligence
The dynamic of belief in the transferable belief model and specialization-generalization matrices

UAI '92 Proceedings of the eighth conference on Uncertainty in Artificial Intelligence
The nature of the unnormalized beliefs encountered in the transferable belief model

UAI '92 Proceedings of the eighth conference on Uncertainty in Artificial Intelligence

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper is a self-contained presentation of a method for combining several belief functions on a common frame that is different from a mere application of Dempster's rule. All the necessary results and their proofs are presented in the paper. It begins with a review and explanation of concepts related to the notion of non-normalized mass-function, or gem-function, introduced by P. Smets under the name basic belief assignment [1,6]. Then the link with Dempster's rule of combination is established. Several results in relation with the notion of Dempster specialization matrix are proved for the first time [2]. Based on these results, the method is then presented and a small application is considered.