An Evolutionary Algorithm based Ontology Alignment Extracting Technology

Ontology matching is able to identify correspondences between heterogeneous ontology entities. During ontology matching process, since different ontology matchers do not necessarily find the same correct correspondences, usually several competing matchers are applied to the same pair of entities in order to increase evidence towards a potential match or mismatch. How to select, combine and tune various ontology matchers to obtain the high quality ontology alignment is one of the main challenges in ontology matching domain. Recently, Evolutionary Algorithms (EAs) are appearing as another suitable methodology to determine the optimal aggregating weights for the matchers. However, existing EA based ontology matching approaches regard various ontology matchers as the black boxes, and try to determine the optimal weights to aggregate their output. Ignoring the effects brought about by different entity mapping’s preference on different matchers could significantly reduce the quality of ontology alignment. Moreover, weights tuned in this way could be problem specific, which means they might not be reused in other ontology matching scenarios. In this paper, we present an EA based ontology alignment extracting technology, which can directly extract the ontology alignment from different matchers’ alignments without tuning their aggregating weights. The experiment is carried out on the bibliographic track of OAEI 2016, and the statistical comparisons with three EA based ontology matching approaches show that our approach is effective to match various heterogeneous ontologies.