SWSCF: A Semantic-based Web Service Composition Framework

Abstract

Web service composition is gaining a considerable momentum as an approach to the effective integration of distributed, heterogeneous, and autonomous application. Current composed service processes are mostly generated in the syntactic level and in a manual way. The limitation of the approach is that it does not allow business to dynamic change partners and services. In order to avoid this disadvantage, it is critical to map complex business from abstract process to executable flows according to application domain semantics and dynamic business requirements. In this paper, we first present a semantic-based Web service composition framework called SWSCF based on both semantic process recommendation and semantic matchmaking. Then, we analyze some key technologies including: (1) Introducing hierarchical activity mechanism for dynamic decomposition of business requirements to find a suitable semantic process template; (2) Searching and selecting service chain to match a semantic activity template which is used to specify an abstract activity; (3) Devising heterogeneous message transforming mechanism to eliminate the incompatible message types during generating data flow of an executable process. Furthermore, by comparison with some service composition frameworks, we prove SWSCF to be more flexible and effective.