摘要
基于Web服务之间相互调用的历史信息,通过演绎和统计的方法挖掘服务之间所蕴含的依赖关系,从而构建服务有向图行为模型,该模型以各Web服务为顶点,以服务间用概率形式表示的相关度为边上的权重.作为模型的应用,给出了基于该模型生成服务合成向导的方法.实验结果表明该方法具有一定的可行性及有效性.
This approach is based on the historical information of services invocation. By means of the deductive and statistical methods, the dependencies among associative Web services are derived. Consequently, to describe these semantics, the directed-graph-based behavior model is constructed, in which nodes are respective Web services, and weights on arcs are the probabilistic association degrees between the 2 linked associative services. As the application of this model, the method is presented to generate the guide for services composition. Experimental results verify the feasibility and effectiveness of our methods.
出处
《郑州大学学报(理学版)》
CAS
2007年第4期126-129,133,共5页
Journal of Zhengzhou University:Natural Science Edition
基金
云南省自然科学基金资助项目
编号2005F0009Q
云南大学中青年骨干教师培养计划项目
教育部春晖计划项目
编号Z2005-2-65003
关键词
WEB服务
行为建模
最小不动点
相关度
Web service
behavior modeling
least fixed point
association degree