The Research Of Semantic Web Service Discovery Model Based On Hp2p Network

With the growth of Web services applications and the improvement in real-time, the centralized registry and discovery mechanism, suffering from many defects such as system bottlenecks, single points of failure, the lower-efficiency of discovering, cannot meet the needs of users. So, distributed web services discovery mechanism is put forward for its distribution and scalability. Also, how to describe Web services to enhance the accuracy of the discovered services has become an active research area.Based on the Semantic Web and P2P Technology, a model for discovery of semantic web services in HP2P-based registries is introduced in this thesis. A new hybrid hierarchical P2P network is used as lower lays for services repository. The network has two layers, in which the upper layer is structured Chord network, and the lower layer is unstructured P2P flooding network. It combines the merits of the unstructured and structured P2P networks for ensuring the stability and expansibility of the model. Further, an ontology parsing algorithm that parses concepts of ontology into a group of simple and sequence value keywords, and an algorithm for annotating Web services based on the value keywords are proposed. Based on them, services annotated by value keywords are published into different nodes so that semantic service clusters in HP2P-based registries are built. Then, a matching algorithm based on value keywords and annotated service characteristics is presented for the implementation of a scalable, efficient and reliable service discovery.