Authors:
Paulo Maio
and
Nuno Silva
Affiliation:
School of Engineering, Polytechnic of Porto, Portugal
Keyword(s):
Ontology, Matching, Mapping, Alignment.
Related
Ontology
Subjects/Areas/Topics:
Applications and Case-studies
;
Artificial Intelligence
;
Knowledge Engineering and Ontology Development
;
Knowledge-Based Systems
;
Ontology Matching and Alignment
;
Symbolic Systems
Abstract:
State-of-the-art ontology matching systems rely on the combination of basic matching techniques but good results are only achieved when processing particular classes of ontologies. Furthermore, they are quite restrictive with respect to their internal configuration, as they are committed to a pre-defined architecture and workflow. Additionally, the skilful selection of matchers and the respective combination and configuration process is difficult and time consuming. Additionally it is hard to test and evaluate. This paper presents a test-bed system that eases the creation of new matching systems. It promotes the reusability, the combination and the configuration of existing matchers, encouraging the development of new matching algorithms able to fill specific open matching gaps exploiting existing methods and algorithms.