Mapping Ontologies to Objects using a Transformation based on Description Logics

Abstract

To manage the increasing complexity of computer systems, a need has arisen to process knowledge instead of only data. Ontologies are nowadays widely used to describe domain knowledge, but although a high level of interest is present with researchers, the technology has not yet sufficiently been put into practice. We present an approach that addresses the transformation of abstract ontological concepts into everyday programming technologies in order to ease the development of semantic web applications for solving common engineering tasks. The presented formal mapping and its implementation -the MOOT framework -is an evolution in the field of ontology to object mapping. They rely on description logics to formalize the transformation process and allow for a detailed discussion about the entailed expressivity. We pay special attention to logical characteristics of roles in order to preserve as much expressivity as possible. Furthermore, an evaluation of the system is presented, where its performance and scalability is demonstrated.