A Granular Space Model for Ontology Learning

Abstract

Ontology learning technology has become a research hotspot in computer science nowadays. The main objective of this paper is to describe domain ontologies at different granularities and hierarchies based on granular computing. A granular space model for ontology learning was explored, and some definitions such as concept granules, granular worlds and the structure of granular space were described formally. Accordingly, the composition and decomposition of concept granules and operation properties were introduced. The proposed model is available for research on ontology learning and data mining at different levels of granularity based on granular computing.