The adoption of ontologies as a formalized approach to information codification is a constant-growing phenomenon in scientific research. Moreover, knowledge sharing and reuse can be improved by adopting hierarchical and modular frameworks and therein embedding available ontologies. Unfortunately, merging procedures may bring about severe, time-consuming problems if a careful selection process is not carried out. Based on these considerations, we propose a methodology for evaluating and selecting higher-level ontologies, given the lower-level ones. Our proposal is based on the computation of ad-hoc metrics which take into account structural and semantic aspects and on the adoption of a multi-decisional analysis procedure. The methodology has been applied to identify ontologies suitable for providing a scientific domain high-level codification to Electromagnetism. Two well-known scientific domain ontologies have been selected and evaluated with the proposed methodology.
展开▼
