On the complexity of the graphical representation and the belief inference in the dynamic directed evidential networks with conditional belief functions

Abstract

Directed evidential graphical models are important tools for handling uncertain information in the framework of evidence theory. They obtain their efficiency by compactly representing (in)dependencies between variables in the network and efficiently reasoning under uncertainty. This paper presents a new dynamic evidential network for representing uncertainty and managing temporal changes in data. This proposed model offers an alternative framework for dynamic probabilistic and dynamic possibilistic networks. A complexity study of representation and reasoning in the proposed model is also presented in this paper. © 2012 Springer-Verlag.