A Framework for Ontology Usage Analysis

Abstract

The Semantic Web (also known as the Web of Data) is growing rapidly and becoming a decentralized social and knowledge platform for publishing and sharing information. In the early days of the Semantic Web (1999-2006), research efforts of the community were centered around knowledge representation; thus, most of the research work was focused on building ontologies (ontology engineering), developing formal languages to represent them (ontology language), methodologies to evaluate and evolve ontologies (ontology evaluation and evolution (OE)), and logic for reasoning with them. As a result of this, even though ontologies were being developed but their instantiation was inadequate to provide the actual instance data needed for the evaluation and analysis of the developed ontologies. In order to overcome this issue, test data was often used to perform the above tasks [1]. However, in the recent past, the focus has shifted towards publishing data either with little or no use of ontologies [2]. This shift in focus is credited to the Linked Open Data (LOD) Project which has published billions of assertions on the Web using well known Linked Data principles. Because of this, the research focus has shifted from knowledge-centered to data-centered and is now settling down at the point where domain ontologies are being used to publish real-world data on the Web. This trend promotes consistent and coherent semantic interoperability between users, systems and applications. In this regard, several domain ontologies have been developed to describe the information pertaining to different domains such as Healthcare and Life Science (HCLS), governments, social spaces, libraries, entertainment, financial service and eCommerce.