Abstract
Ontology is a new paradigm introduced with the semantic web to describe in an explicit and formal way the various aspects of knowledge of a specific field. For this purpose, a single ontology may not be comprehensive to represent all due to the lack of a common and shared ontology between communities. Ontologies need to establish a number of interlinks to ensure communication between them, which is not always obvious because of their terminological, syntactic and semantic heterogeneity.
The proposed matching system aims to discover in an automatic way, the correspondence links between two intrinsically heterogeneous ontologies, through different techniques of calculations of similarity between their entities. It allows to reveal on one hand the issue of searching for the most relevant, coherent and meaningful alignments and on the other hand, to propose a new strategy that ensures flexibility and scalability of the system by the combination of the matchers.