Interactive Cross-Lingual Ontology Matching

Recently, with the growing number of ontologies defined in different languages, to bridge the semantic gaps between them, it is necessary to identify the correspondences between their heterogeneous entities, so-called cross-lingual ontology matching. Due to the complexity and the intricacy of the cross-lingual ontology matching, it is essential to get an expert involved in the matching process to guarantee the alignment's quality. In this paper, we propose an interactive cross-lingual ontology matching technique that makes the user and automatic matcher work together to create high-quality alignments in a reasonable amount of time. In particular, we present a cross-lingual similarity metric to calculate the similarity value of two cross-lingual entities, construct an optimal model for the cross-lingual ontology matching problem, and propose an interactive compact differential evolution (ICDE) algorithm to effectively match the cross-lingual ontologies. The experiment exploits the ontology alignment evaluation initiative (OAEI) multifarm track to test our proposal's performance. The experimental results show that the ICDE significantly outperforms other EA-based matchers and OAEI's participants, and the interacting mechanism can significantly improve the alignment's quality.