Choosing the Granularity of Abstraction Networks for Orientation and Quality Assurance of the Sleep Domain Ontology

An abstraction network is a compact network summarizing the structure and content of a given ontology. Abstraction networks have been shown to support orientation into and quality assurance of ontologies. Area and partial-area taxonomies are examples of abstraction networks that utilize the relationships of an ontology to group together classes with similar structure and semantics. These taxonomies can be derived in different ways, leading to different granularities of summaries. Such granularity is illustrated by applying various derivation methodologies to the Sleep Domain Ontology (SDO), hosted on BioPortal. The impact of different granularity levels is demonstrated with respect to orientation into and quality assurance of the ontology’s structure and content.