由于Web服务的数量日渐增长,所以快速准确地发现满足用户需求的高质量服务需要基于语义和服务质量的分布式服务发现架构的支持。通过将使用非负实数向量表示的服务描述发布到具有特定服务类型支持的P2P注册结点并配合使用Bloom Filter、本体概念组和预先计算出的匹配信息,能降低系统系统查找满足用户请求的服务的代价。采用对查询结果进行评级和用户将所用服务的质量信息反馈给注册结点的技术能为用户找到更高质量的服务,增加了系统工作的精确性和灵活性。
The growing number of Web services advocates distributed discovery infrastructures which are semantics- enabled and support quality of services (QoS). In this paper, service descriptions are converted to service characteristic vectors according to mappings of ontological concepts to non- negative numerical key values and ontological concepts groups composed of the concepts with the approximately same semantics. Then vectorized service descriptions are advertised m certain P2P registries each of which manages certain type of services with the similar service characteristic vector. During discovery process using Bloom Filter and pre - computed matching information enable minimizing search efforts. It is possible for the system to provide the high quality services by using the technologies that rank the query results and provide feedhaeks to P2P registry responsible for the management of the service.