Artificial Intelligence


Evidential Divergence Measures in Dempster–Shafer Theory

Authors: Fuyuan Xiao

The Dempster–Shafer evidence (DSE) theory, as a generalization of the Bayes probability theory, has more capability to handle the uncertainty in the decision-making problems. In the DSE theory, however, how to measure the divergence between basic belief assignments (BBAs) is still an open issue which has attracted many attentions. On account of this point, in this paper, new evidential divergence measures are developed to measure the difference between BBAs in the DSE theory, called as EDMs. The EDMs consider both of the correlations between BBAs and the subset of set of BBAs, respectively. Consequently, they can provide a much more convincing and effective way to measure the discrepancy between BBAs. In a word, the EDMs as the generalization of the divergence measures in the Bayes probability theory have the universal applicabilities. Additionally, a new Belief–Jensen–Shannon divergence measure is derived based on the EDMs, in which different weights can be assigned to the BBAs involved, so that it provides a promising solution to be applied in solving the problems of decision-making. Finally, numerical examples are illustrated that the proposed methods are more feasible and reasonable to measure the divergence between BBAs in the DSE theory.

Comments: 3 Pages.

Download: PDF

Submission history

[v1] 2019-03-28 21:11:52

Unique-IP document downloads: 9 times is a pre-print repository rather than a journal. Articles hosted may not yet have been verified by peer-review and should be treated as preliminary. In particular, anything that appears to include financial or legal advice or proposed medical treatments should be treated with due caution. will not be responsible for any consequences of actions that result from any form of use of any documents on this website.

Add your own feedback and questions here:
You are equally welcome to be positive or negative about any paper but please be polite. If you are being critical you must mention at least one specific error, otherwise your comment will be deleted as unhelpful.

comments powered by Disqus