Seminář: Designing and visualizing ontologies – and how could ontological background models help

In order to publish structured, machine-readable data on the semantic web, ontologies and vocabularies (expressed in the OWL language) are needed, which define the concepts by which the data are annotated. Design and usage of the ontologies may be eased by ontology visualization tools; many such tools nowadays exist, which rely on diverse ontology displaying principles. The first part of the presentation will demonstrate a recommender system (namely, a rule base for the NEST expert system shell) designed to suggest the most suitable tool based on ontology characteristics and visualization use case scenario. Yet, deciding which ontology to use for a specific data annotation case might be difficult even with the best available visualization tool. The second part of the presentation will therefore present a possible way of analyzing and comparing the coverage of multiple ontologies from the same domain, based on visualizing a so-called PURO ontological background model (a kind of instance-level meta-model for ontologies). Finally, the third part of the presentation will outline how ontological background models might support a revolutionary way of ontology authoring, which borrows some ideas from relational database design.