An Effective Triangulation Algorithm for Bayesian Network Inference
Zhang Con · 2005
Bayesian network is a powerful tool on non-deterministic knowledge representation and inference. The ex- act inference of Bayesian network is a NPC problem, the main difficulty is to triangulate the network and construct a join tree with minimum weight. This work presents a new triangulation algorithm: MsLB-Triang. MsLB-Triang is based on both the Direc property and LB-simple property of triangulation graph, and outperforms the popular using Min. Weight Heuristic triangulation algorithm at the total weight and total filling edges of result graph.