Abstract: The paper proposes a technique based on semantic Web. Firstly, a service ontology should be defined. The service providers descript service information by the means of the service ontology and distribute it in the form of semantic Web. Users can define queries to find services they need. Queries can be defined in various forms, such as semantic Web and three-triples. It needs an algorithm to handle the users' queries, which can understand semantic information and do inferences over it. The algorithm matches the users' queries and the services provided, and lastly returns the results. The paper analyzes inference mechanism of semantic Web and proposes an algorithm for fast and exact service discovery.

Key words: Service discovery; Semantic Web; Makeup language; Semantics; Inference; Match algorithm

摘要： 提出了基于语义Web 的技术。首先要定义一个服务本体，服务提供者通过服务本体描述服务信息，并以语义Web 的形式来发布。用户可以定义查询条件来查询自己所需要的服务。查询条件可以多种形式定义，如语义Web 形式和三元组形式。这需要有一个算法来处理用户查询，能够处理语义信息并进行推理，对用户的查询和所提供的服务进行匹配，并返回查询结果。文章研究了语义Web 的推理机制，并提出了一种用于快速和准确的服务发现的匹配算法。

关键词: 服务发现；语义Web；标志语言；语义；推理；匹配算法