Generating Bayesian Networks from Probability Logic Knowledge Bases
Peter Haddawy Department of Electrical Engineering and Computer Science University of WisconsinMilwaukee PO Box 784 Milwaukee, WI 53201
We present a method for dynamically gen erating Bayesian networks from knowledge bases consisting of firstorder probability logic sentences. We present a subset of proba bility logic sufficient for representing the class of Bayesian networks with discretevalued nodes. We impose constraints on the form of the sentences that guarantee that the knowl edge base contains all the probabilistic infor mation necessary to generate a network. We define the concept of dseparation for knowl edge bases and prove that a knowledge base with independence conditions defined by d separation is a complete specification of a probability distribution. We present a net work generation algorithm that, given an in ference problem in the form of a query Q and a set of evidence E, generates a network to compute P (Q|E). We prove the algorithm to be correct.