Modeling and Reasoning about Changes in Ontology Time Series

Ontologies evolve when the underlying domain world changes at different points of time. The result then is a series of ontologies whose concepts are related with each other not only within one ontology valid at a moment but through the time, too. This chapter presents a model for representing ontology time series. The focus is on modeling partial overlap between concepts evolving over long periods of time, and the domain of application is historical geospatial reasoning. A framework is presented for representing and reasoning about conceptual overlap of concepts that evolve over an ontology time series. The idea is to provide the ontology developer with an intuitive change ontology for expressing local ontological changes in a declarative way. An algorithm is presented for reasoning about overlapping concepts globally over long periods of time. This algorithm can be applied, e.g., in concept-based information retrieval for ranking search results according to their relevance. ONTOLOGY CHANGE AND INFORMATION RETRIEVAL Ontologies are a key technology underlying the Semantic Web (Fensel, 2004). They are used for defining vocabularies by which the metadata describing web contents is represented in a machine-interpretable way. Based on ontologies, intelligent content-based web services can be created and semantic interoperability of web systems enhanced. ∗Tomi Kauppinen and Eero Hyvönen. Modeling and Reasoning about Changes in Ontology Time Series. Published as a peer reviewed chapter (chapter 11) in Book: Ontologies: A Handbook of Principles, Concepts and Applications in Information Systems. Rajiv Kishore, Ram Ramesh, Raj Sharman (editors). SpringerVerlag, New York, 2007. 1