Ontologies and vocabularies are commonly used within Linked data cloud for typing individual objects. Due to their semantics, domain ontologies are used as a source of knowledge in diverse applications. Data mining phases are commonly enhanced by various forms of background knowledge, besides other things, to explain discovered patterns. In this poster paper we propose how to enhance a selection of proper explanations for data mining results based on ontologies. The approach of offering relevant explanations is described and demonstrated on an example. The poster paper is wrapped up with discussions, conclusions and future work.
Department of Information and Knowledge Engineering, University of Economics, Prague
https://keg.vse.cz/
