Adaptive ontology re-use: finding and re-using sub-ontologies

Purpose – The discovery of the “right” ontology or ontology part is a central ingredient for effective ontology re-use. The purpose of this paper is to present an approach for supporting a form of adaptive re-use of sub-ontologies, where the ontologies are deeply integrated beyond pure referencing. Design/methodology/approach – Starting from an ontology draft which reflects the intended modeling perspective, the ontology engineer can be supported by suggesting similar already existing sub-ontologies and ways for integrating them with the existing draft ontology. This paper's approach combines syntactic, linguistic, structural and logical methods into an innovative modeling-perspective aware solution for detecting matchings between concepts from different ontologies. This paper focuses on the discovery and matching phase of this re-use process. Findings – Owing to the combination of techniques presented in this general approach, the work described performs in the general case as well as approaches tailored for a specific usage scenario. Research limitations/implications – The methods used rely on lexical information obtained from the labels of the concepts and properties in the ontologies, which makes this approach appropriate in cases where this information is available. Also, this approach can handle some missing label information. Practical implications – Ontology engineering tasks can take advantage from the proposed adaptive re-use approach in order to re-use existing ontologies or parts of them without introducing inconsistencies in the resulting ontology. Originality/value – The adaptive re-use of ontologies by finding and partially re-using parts of existing ontological resources for building new ontologies is a new idea in the field, and the inclusion of the modeling perspective in the computation of the matches adds a new perspective that could also be exploited by other matching approaches.