Abstract
Ontology mapping is the process whereby semantic relations are defined between two ontologies at conceptual level, which in turn are applied at data level transforming source ontology instances into target ontology instances. Besides database schema mapping is a well-known problem, ontology mapping, as a similar problem, faces new challenges in the context of Semantic Web, especially concerning higher level of heterogeneity, evolution, distribution, autonomy and limitations on representation technology. This paper focuses on the formalization, representation, specification and execution of ontology mapping. The architecture of the system is based on the notion of Multi-dimensional Service, representing not only the system transformation capabilities, but also the expertise in the manipulation of specific semantic relations. In fact, the process of choosing, applying, validating, evolving or negotiating a semantic relation depends on both the ontologies contents and the transformations available in the system. This new architecture and mapping process are being tested and validated in MAFRA Toolkit, a specific implementation of MAFRA. The main contributions of this paper are the formalization of the ontology mapping process, the specification and execution methodology and the multi-dimensional service-oriented architecture.
Get full access to this article
View all access options for this article.