Constraint-Based Bayesian Network Structure Learning using Uncertain Experts’ Knowledge

Exploiting experts' knowledge can significantly increase the quality of Bayesian network (BN) structures produced by learning algorithms. However, in practice, experts may not be 100% confident about opinions they provide. Worst, latter also conflicting. Including such specific algorithms is therefore complex. In literature, there exist a few score-based that exploit both data and existence/absence arcs BN. But, as far we know, no constraint-based algorithm capable exploiting knowledge. this paper, fill gap introducing mathematical foundations for new independence tests including kind information. We provide relying on these well experiments highlight robustness our method its benefits compared to other