A Probabilistic Framework for Information Integration and Retrieval on the Semantic Web
Livia Predoiu, Heiner Stuckenschmidt · MADOC (University of Mannheim) · 2007
We present a framework for probabilistic Information Processing on the Semantic Web that is capable of representing ontologies, deductive databases, uncertain mappings between them, results of statistical instance classification and ontology learning. Our framework is built on a knowledge representation formalism called Bayesian Description Logic Programs because it is a probabilistic extension of Description Logic Programms and a fragment of Bayesian Logic Programs. We show in this paper how to perform information integration and retrieval within our framework by means of integrated reasoning.