An efficient semantic web services matching mechanism

1Citations
Citations of this article
8Readers
Mendeley users who have this article in their library.
Get full text

Abstract

Bringing semantics to Web Services description and matching remarkably improves the precision and recall performance of service discovery. However, it greatly increases the service matching time by executing semantic reasoning, and makes the real-time service selection and composition harder than before. To shorten service matching time, this paper presents an efficient Semantic Web Services matching mechanism with semantic information pretreatment. It shifts part of the reasoning process from service matching phase to service publishing phase. A prototype of Semantic Web Services matchmaker has been implemented based on this mechanism. The experiments on two service datasets show that our mechanism outperforms existing methods. © 2010 Springer-Verlag.

Cite

CITATION STYLE

APA

Li, J., Ma, D., Zhao, Z., Zhu, H., & Li, L. (2010). An efficient semantic web services matching mechanism. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 6162 LNCS, pp. 110–119). https://doi.org/10.1007/978-3-642-14415-8_8

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free