Ontology alignment based on Multi-Agents Systems

Abstract

The ontology alignment is based on the principle of comparison to link entities together. This is usually done manually by experts. However, the increase of ontologies, in number and size, need urgently tools development, to assure an automatic/semiautomatic alignment. The Multi Agents Systems (MAS) paradigm is known by its means in communication, cooperation and negotiation for efficiency in terms of cost and execution time. In this paper, we propose an approach to ensure an automatic ontology alignment by introducing a new measure of linguistic similarity and using the MAS paradigm. An evaluation of the proposed approach, with and without MAS, is made with the Ontology Alignment Evaluation Initiative (OAEI). We will show that the use of MAS is more efficient in terms of execution time.