A Comparative Evaluation of String Similarity Metrics for Ontology Alignment
Yufei Sun · Journal of Information and Computational Science · 2015
Ontology alignment is regarded as the most perspective way to achieve semantic interoperability among heterogeneous data. The majority of state of art ontology alignment systems used one or more string similarity metrics, while the performance of these metrics were not given much attention. In this paper we flrst analyze naming variations in competing ontologies, then we evaluate a wide range of string similarity metrics, from the experimental result we can get some heuristic strategies to achieve better alignment results with regard to efiectiveness and e‐ciency.