基于本体概念语义的Web服务发现

Web service discovery based on ontology concept semantics

为了提高服务发现的准确性,改善服务自动选择、组合及执行的效率,通过对传统的语法或表面语义进行服务匹配所产生结果的分析,发现错误的肯定和错误的否定匹配结果源自缺少对于服务广告和目标的意图语义描述.使用本体概念关系描述服务的广告和目标,通过概念所代表的具体对象集合的范围描述服务提供者和请求者的使用意图.实验结果表明:考虑服务使用意图的匹配结果细粒度划分,能够将服务发现结果对应到4类情况之一,结合传统的服务匹配算法,可以得到20种匹配结果.对于给定的服务数据集中的服务广告和目标增加与其意图相关的上下文信息,该方法能够显著地提高服务发现的准确度.

In order to increase the accuracy performance for service discovery and improve the efficiency of service selection, composition and execution in automation way, the outcomes from traditional syntax or superficial semantics matching algorithm are analyzed. It is found that the false positive and the false negative of service matching results are derived from lack of intention semantics description for service advertisement and goal. Service advertisement and goal are described by using the relationship of ontology concepts; the intention of service provider and consumer is described by using the range of instance object sets ascribed to a certain concept. Experiments show that service discovery results can be corresponding to one of four different cases by considering well granular partition for matching results of service usage intention. In case of combination with traditional matching methods, twenty kinds of results can be acquired. If context information in relation with the user intention is appended to the service advertisement and goal in given service data sets, the approach can increase the accuracy degree of service discovery significantly.