Efficient Ontology Meta-Matching Based on Metamodel-assisted Compact MOEA/D

In this paper we propose an improved Multi-Objective Evolutionary Algorithm based on Decomposition (MOEA/D), i.e. metamodel-assisted compact MOEA/D, to efficiently solve the ontology meta-matching problem. Particularly, we dedicate to determine the optimal ontology alignment by tuning four different basic similarity measures, i.e. Syntactic Measure, Linguistic Measure, Taxonomy-based Measure and Instance-based Measure. The experimental results show that our proposal can significantly improve the efficiency of MOEA/D based approach without sacrificing the quality of the ontology alignment, and the results obtained by our approach are better than state-of-the-art ontology matching systems.