Abstract

Ontologies allow the capture and sharing of domain knowledge by formalizing information and making it machine understandable. As part of an information system,
ontologies can capture and carry the reasoning knowledge needed to fulfill different application goals. Although many ontologies have been developed over recent years, few
include such reasoning information. As a result, many ontologies are not used in real-life applications, do not get reused or only act as a taxonomy of a domain. This work is an investigation into the practical use of ontologies as a driving factor in the development of applications and the incorporation of Knowledge Engineering as a meaningful activity into modern agile software development. This thesis contributes a novel methodology that supports an incremental requirement analysis and an iterative formalization of ontology design through the use of ontology reasoning patterns. It also provides an application model for ontology-driven applications that can deal with nonontological data sources. A set of case studies with various application specific goals helps to elucidate whether ontologies are in fact suitable for more than simple knowledge formalization and sharing, and can act as the underlying structure for developing largescale information systems. Tasks from the area of bug-tracker quality mining and clone detection are evaluated for this purpose.