Computational-workload based binarization and partition of qualitative Markov trees for belief combination

Abstract

Binary join trees have been a popular structure to compute the impact of multiple belief functions initially assigned to nodes of trees or networks. Shenoy has proposed two alternative methods to transform a qualitative Markov tree into a binary tree. In this paper, we present an alternative algorithm of transforming a qualitative Markov tree into a binary tree based on the computational workload in nodes for an exact implementation of evidence combination. A binary tree is then partitioned into clusters with each cluster being assigned to a processor in a parallel environment. These three types of binary trees are examined to reveal the structural and computational differences.