Enabling Web Service Discovery in Heterogeneous Environments

Abstract

In large, open environments, service discovery has to face the challenge of heterogeneity. Service advertisements published by different organizations or individuals may differ in their description models, thus expressiveness levels. Even on the same expressiveness level, organizations may use assorted domain ontologies. In addition, a service discovery tool may not have a global view of all service advertisements in the system. Unfortunately, most contemporary service discovery approaches rely on these key factors. This paper presents a method that addresses the mismatch problem in description models. A neutral model is proposed in this work for aligning different service description models. A matchmaking method that encompasses different semantic, syntactic and hybrid service description languages based on this neutral model is also presented. Implementation and evaluation of the proposed method showed a satisfactory result.