This paper investigates the problem of belief update in Bayesian networks (BN) with uncertain evidence. Two types of uncertain evidences are identified: virtual evidence (reflecting the uncertainty one has about a reported observation) and soft evidence (reflecting the uncertainty of an event one observes). Each of the two types of evidence has its own… (More)
This paper presents an efficient method, SMOOTH, for modifying a joint probability distribution to satisfy a set of inconsistent constraints. It extends the well-known "iterative proportional fitting procedure" (IPFP), which only works with consistent constraints. Comparing with existing methods, SMOOTH is computationally more efficient and insensitive to… (More)
Previously w e have proposed a theoretical framework, called BayesOWL, to model uncertainty in semantic web ontologies based on Bayesian networks. In particular, we have developed a set of rules and algorithms to translate an OWL taxonomy into a BN. In this paper, we describe our implementation of BayesOWL framework together with examples of its use.
This paper presents a formal convergence proof for E-IPFP, an algorithm that integrates low dimensional probabilistic constraints into a Bayesian network (BN) based on the mathematical procedure IPFP. It also extends E-IPFP to deal with constraints that are inconsistent with each other or with the BN structure. is the set of conditional probability tables… (More)
Previously we have proposed a theoretical framework, named BayesOWL, which translates an OWL taxonomy of concept classes into a Bayesian network (BN) and incorporates consistent probabilistic information about the concept classes into the translated BN. In this paper, we extend the original framework to support general OWL DL ontologies and to effectively… (More)