Ontology alignment using multiple contexts
Jeffrey Partyka, Neda Alipanah, Latifur R. Khan, Bhavani M. Thuraisingham, Shashi Shekhar · CEUR Workshop Proceedings · 2008
Ontology alignment involves determining the semantic heterogeneity between two or more domain specifications by considering their associated concepts. Our approach considers name, structural and content matching techniques for aligning ontologies. After comparing the ontologies using concept names, we examine the instance data of the compared concepts and perform content matching using value types based on N-grams and Entropy Based Distribution (EBD). Although these approaches are generally sufficient, additional methods may be required. Subsequently, we compare the structural characteristics between concepts using Expectation-Maximization (EM). To illustrate our approach, we conducted experiments using authentic geographic information systems (GIS) data and generate results which clearly demonstrate the utility of the algorithms while emphasizing the contribution of structural matching.