Integrating Trust with User Preference for Effective Web Service Composition

摘要

Web service composition is a process to compose homogenous or heterogeneous services together in order to create value-added services. Many non-functional features including QoS and user preferences have been adopted to guide such a process. However, two issues are observed: (1) the expressiveness of user preference is subject to quantitative preferences without proper use of qualitative preferences; (2) a highly preferred composite service may not be trustworthy, or a highly trustworthy composite service may not be preferable. The existing studies concentrate on either user preference or service trust, and fail to provide a systematic solution to integrate both user preference and service trust together for service compositions. To address these issues, we combine both qualitative and quantitative preferences as well as service trust together in the process of service composition. We investigate the application of heuristic algorithms on multi-objective optimization for the service composition problem. A new hybrid nature inspired intelligent algorithm is also proposed and compared with other popular heuristic algorithms. We aim to obtain optimal web service compositions that can satisfy these (potentially conflicting) constraints as much as possible. Results demonstrate the efficiency and effectiveness of our approach in comparison with other counterparts.