Decision support in e-business based on assessing similarities between ontologies

摘要

As ontology usage becomes more prevalent in e-business decision support systems, it is essential to establish interoperation between agents or web services with different ontologies. Ontology matching represents a promising technique in information integration, agent communication and web services. Here, an ontology similarity assessment algorithm is proposed to map concepts and properties between ontologies of semantic web services. It has the two phases (1) syntactic analysis based on string tokenisation, and (2) semantic analysis employing the WordNet thesaurus and tree-structured graphs with the Tversky model. Promising experimental results indicate that this algorithm can find matching results with relatively high precision.