服务失效情境下的高质量Web服务推荐

Recommendation of High-quality Web Service in the Failure Context

推荐有效且高质量的Web服务是Web服务技术应用的关键.由于服务所处的环境动态变化,现有推荐方法推荐的服务常存在不可用的情况.文中提出一种服务失效情境下高质量Web服务推荐方法.基于Web服务的历史使用信息,将Web服务及服务之间的关系抽象成一个二部图来为用户推荐合适的Web服务.若推荐的服务失效,则采用基于Word Net的概念语义进行功能相似度匹配,然后根据服务的调用率,从Web服务描述文档质量角度构建Web服务质量的评价模型,从而选取服务来替换无效的Web服务,向用户推荐更高质量更为可用的服务.使用my Experiment上的工作流服务和Web服务的真实数据来说明本文方法的有效性,实验表明该方法能够有效地实现推荐服务的兼容性和可用性.

Web Service recommendation is the key for the application of the Web service technology. The current service recommenda- tion approaches usually cannot be applied to the dynamically changing environment. This paper proposes a recommendation method for high-quality web services when the recommended service is invalid. First, this method recommends appropriate web services to users through deriving a bipartite graph from the relationships between web services based on the history of services composition. If the rec- ommended service is invalid,conceptual semantics based on WordNet are adopted to get the similarity of function between different services. Then,invalid service is replaced by the service which is selected based on a quality evaluation model. The quality evaluation model is built to reflect the qualifies of web services from the perspective of web services documents. Finally, workflow services and web services in myExperiment are used to demonstrate the effectiveness of the proposed method. Related experiment illustrate the method can assure the compatibility and availability of the recommended services.