Web服务合成是目前的研究热点之一,是Web服务架构中的重要组成部分.Web服务合成过程模型用以描述Web服务合成中各服务组件应该满足的语义约束信息,是自动服务发现及合成的基础.以事件代数作为支撑理论,旨在为Web服务合成过程建模提供一种支持语义表示及推理的形式化方法.针对Web服务合成的特点及合成过程本身具有的不确定性,首先基于统计学习的思想对传统事件代数进行不确定性扩展,给出@eb服务不确定事件代数(WSUEA)的定义、操作及语义,进而对服务合成中的顺序、分支及并行等基本情形分别给出其WSUEA表达式、以及相应的概率计算方法.实验验证了所提出方法的有效性.
Web services composition is widely studied and hotly debated in recent years, and it is one of important components of the Web services architecture. The process model of Web services composition is used to describe the semantic constraints that should be satisfied among service components, and is the basis for automatic service discovery and composition. Taking event algebra as the underlying theory, it is to provide process modeling of Web services composition with a formal approach supporting representation and reasoning of semantics. Aiming at the characteristics and inherent uncertainties implied by services composition itself, it is first extended the classical event algebra and given the definitions, operations and semantics of Web service uncertain event algebra(WSU EA). Further, it is given give the formal WSU EA expressions for sequential, conditional and parallel cases in services composition, as well as the method for computing corresponding probabilities. The feasibility of our methods is verified by experiments.