DocumentCode
59019
Title
Dependence-Guided Service Composition for User-Centric SOA
Author
Wei-Tek Tsai ; Peide Zhong ; Xiaoying Bai ; Elston, Jack
Author_Institution
Sch. of Comput., Inf., & Decision Syst. Eng., Arizona State Univ., Tempe, AZ, USA
Volume
8
Issue
3
fYear
2014
fDate
Sept. 2014
Firstpage
889
Lastpage
899
Abstract
Service-oriented architecture (SOA) is characterized by dynamic service discovery and composition. For user-centric SOA, not only services, workflows, and application templates can be published and discovered for composition. Furthermore, a user can publish an application template and let providers supply the needed services. A key issue in user-centric service composition is to intelligently and effectively discover the subset of correlated services that best match the users´ requirements. This paper proposes a two-step composition process. In the first step, users choose templates from ontology with its dependence. The dependence information identifies a set of candidate services and workflows that may be applicable for composition. In the second step, after the templates are selected, users can finalize the selection of services and workflows from a set of candidate services or workflows based on their preference. This two-step process is supported by dependence identification algorithms and is illustrated using a case study.
Keywords
formal specification; ontologies (artificial intelligence); service-oriented architecture; application template; dependence identification algorithm; dependence information; dependence-guided service composition; dynamic service composition; dynamic service discovery; ontology; service selection; service-oriented architecture; service-oriented specification; template selection; two-step composition process; user requirements; user-centric SOA; user-centric service composition; workflow selection; Business; Collaboration; Ontologies; Service-oriented architecture; Unified modeling language; Dependence; ontology; service composition;
fLanguage
English
Journal_Title
Systems Journal, IEEE
Publisher
ieee
ISSN
1932-8184
Type
jour
DOI
10.1109/JSYST.2013.2260947
Filename
6637058
Link To Document