A novel ontology matching technology based on NSGA-II

Ontology is constructed or researchers to overcome the heterogeneous problem in a domain, but merely using ontology may raise the heterogeneous problem to a higher level. To solve the heterogeneous problem between two ontologies, it is necessary to determine the relationships that hold between the entities in them. The process of finding these correspondences is called ontology matching and the matching results are called ontology alignment. Various ontology matching approaches have been proposed so far, and the Evolutionary Algorithm (EA) based ontology matching technologies have been attracting more and more attentions, although the quality of the alignments obtained and the efficiency of the algorithm are both barely satisfactory. To address these issues in EA based ontology matching technologies, in this paper, an novel ontology matching technology based on NSGA-II is presented. In particular, in our work, a novel similarity measure based on Information Theory and a special mapping extraction approach based on the Naive Descending Extraction (NDE) algorithm are respectively proposed, a Multi-objective optimal model for ontology matching problem is presented and the problem-specific NSGA-II is designed. Experimental results show that our proposal is efficient and can find the best solution so far. © 2016.