A Layered Approach for Merging Application Ontologies: A Case Study from Agriculture Domain

Ontology is a machine interpretable way of representing knowledge in a precise and complete way favorable to solve many problems in the field of knowledge engineering. Different knowledge areas evolve with the time and the applications which use ontologies should be updated with new knowledge accordingly. It is more effective to combine the ontologies with new knowledge with the existing application ontologies rather than designing a new ontology from scratch. When combining ontologies, keeping the original usability of the initial ontologies and the heterogeneity of the components of the ontologies are the main obstacles. As there are no universal standard for naming of ontology components, this is a major reason for the heterogeneity problem. Methods for overcoming these problems are needed. In this research, we have proposed two algorithms to overcome the aforementioned problems. These algorithms for finding correspondences of ontology concepts and merging domain specific application ontologies keeping the original usability of the initial ontologies are the main outcomes of this research. The proposed algorithms are evaluated in terms of accuracy by comparing the resultant ontology merged using the proposed algorithm and the resultant ontology merged by an expert. The evaluation results prove that the proposed methodology merges the domain specific application ontologies very similar to the ontology merged by human intervention.