MARS Collection: C4I Papers

Belief in Belief Functions: An Examination of Shafer's Canonical Examples

Sun, 01 Jan 1989 00:00:00 GMT

Title: Belief in Belief Functions: An Examination of Shafer's Canonical Examples

Authors: Laskey, Kathryn B.

Abstract: In the canonical examples underlying Shafer-Dempster theory, beliefs over the hypotheses of interest are derived from a probability model for a set of auxiliary hypotheses. A belief function differs from a Bayesian probability model in that one does not condition on those parts of the evidence for which no probabilities are specified. The significance of this difference in conditioning assumptions is illustrated with two examples giving rise to identical belief functions but different Bayesian probability distributions.

Credibility Models for Multi-Source Fusion

Sat, 01 Jul 2006 00:00:00 GMT

Title: Credibility Models for Multi-Source Fusion

Authors: Wright, Edward J.; Laskey, Kathryn B.

Abstract: This paper presents a technical approach forfusing information from diverse sources. Fusion requiresappropriate weighting of information based on thequality of the source of the information. A credibilitymodel characterizes the quality of information based onthe source and the circumstances under which theinformation is collected. In many cases credibility isuncertain, so inference is necessary. Explicitprobabilistic credibility models provide a computationalmodel of the quality of the information that allows use ofprior information, evidence when available, andopportunities for learning from data. This paperprovides an overview of the challenges, describes theadvanced probabilistic reasoning tools used to implementcredibility models, and provides an example of the use ofcredibility models in a multi-source fusion process.

A Proposal for a W3C XG on Uncertainty Reasoning for the World Wide Web

Wed, 01 Nov 2006 00:00:00 GMT

Title: A Proposal for a W3C XG on Uncertainty Reasoning for the World Wide Web

Authors: Laskey, Kenneth J.; Costa, Paulo C. G.; Laskey, Kathryn B.

Abstract: The Semantic Web envisions effortless cooperation between humans andcomputers, seamless interoperability and information exchange among web applications,and rapid and accurate identification and invocation of appropriate Web services. At thecurrent stage of evolution in Semantic Web research, there is a growing understandingthat a major step towards this vision involves the implementation of principled uncertaintyrepresentation and reasoning in SW applications. This position paper introduces initialthoughts on how the World Wide Web Consortium (W3C) Incubator XG process could beemployed to move forward the concept of a Web with uncertainty.

Probabilistic Ontologies for Efficient Resource Sharing in Semantic Web Services

Wed, 01 Nov 2006 00:00:00 GMT

Title: Probabilistic Ontologies for Efficient Resource Sharing in Semantic Web Services

Authors: Costa, Paulo C. G.; Laskey, Kathryn B.; Laskey, Kenneth J.

Abstract: Service Oriented Architecture (SOA) is a key technology to supportinteroperability among data and processing resources. Semantic interoperabilityrequires mapping between vocabularies of independently developed resources,a task fraught with uncertainty. Probabilistic ontologies enable representation ofknowledge in domains characterized by uncertainty. As such, they promise toimprove the quality of service descriptions, enable more thorough analysis ofservice composition opportunities, and provide a theoretically sound methodologyfor semantic mapping under uncertainty. This paper defines probabilisticontologies, discusses their application to SOA, and presents a conceptualscheme for using a federation of ontologies (with both common and probabilisticontologies) as a semantic mapping tool for service oriented information exchangesystems with different levels of service descriptions (including legacyand probabilistic enabled descriptions).

PR-OWL: A Framework for Bayesian Ontologies

Wed, 01 Nov 2006 00:00:00 GMT

Title: PR-OWL: A Framework for Bayesian Ontologies

Authors: Costa, Paulo C.G.; Laskey, Kathryn B.

Abstract: Across a wide range of domains, there is an urgent need for a wellfoundedapproach to incorporating uncertain and incomplete knowledge into formaldomain ontologies. Although this subject is receiving increasing attentionfrom ontology researchers, there is as yet no broad consensus on the definition of aprobabilistic ontology and on the most suitable approach to extending current ontologylanguages to support uncertainty. This paper presents two contributions todeveloping a coherent framework for probabilistic ontologies: (1) a formal definitionof a probabilistic ontology, and (2) an extension of the OWL Web OntologyLanguage that is consistent with our formal definition. This extension, PR-OWL,is based on Multi-Entity Bayesian Networks (MEBN), a first-order Bayesian logicthat unifies Bayesian probability with First-Order Logic. As such, PR-OWL combinesthe full representation power of OWL with the flexibility and inferentialpower of Bayesian logic.