Light-weight cross-lingual ontology matching with LYAM++

Abstract

During the last decade, several automatic ontology matching systems were developed to address the problem of ontology heterogeneity. Aligning cross-lingual ontologies is among the current challenging issues in the field of ontology matching. The majority of the existing approaches rely on machine translation to deal with this problem. However, inherent problems of machine translation are imprecision and ambiguity. In this paper, we propose a novel approach to the cross-lingual ontology matching task, relying on the large multilingual semantic network Babel Net as a source of background knowledge to assist the matching process. We have designed and tested a novel orchestration of the components of the matching workflow. Our approach is implemented under the form of a prototype named LYAM++ (Yet Another Matcher–Light)—a fully automatic cross-lingual ontology matching system that does not rely on machine translation. We report the results of our experiments thatb show that LYAM++ outperforms considerably the best techniques in the state-of-the-art according to the obtained results on the Multi Farm datasets of the Ontology Alignment Evaluation Initiative 2014.