A P2P Network Resource Searching Mechanism Based On The Semantic Theory

The hottest distribution research point at present is the Peer-to-Peer network due to its characteristics: expansibility, fault-tolerant and self-determination. But the challenge in the field of this research is how to carry out the searching service in a highly effective way, especially in the complicated condition of widely distributing users, enormous quantities, node behaviors being out of control and the asymmetric computation capability and network connection. Most of the P2P file searching systems only support simple query like keywords consulting. Their searching accurate ratio is not high as to the fuzzy quering request or multi-keywords quering because of the defects theirselves. And as the resources are lack of general semantic describing, the users are always difficult to find out the relative contents and it is hard to make the resources be semanticly reimmersed.As to the above questions, this thesis overcomes the shortcomings of traditional client/server-based searching mode and improves the system scalability taking advantage of P2P and semantic Web technology. Simultaneously,it labels resources with semantic and combines the described logic ideas, which advances the ability of analyzing and understanding knowledge. For that, the system can be used in some complicated inquiry which is more difficult than key-word inquiry, and finally implements the results with full satisfaction.The thesis has proposed a search mechanism SBSGSS based on the semantic category and P2P network resource of semantic similarity. It has the information model with all the nodes sharing on the basis of the knowledge in the field of the ACM computer document. The information model guarantees the resources understanability in time by semanticly describing the resources, and makes the complicated inquiry and highly sharing knowledges be true. By the way, using the describing logic as the analyzing principle, the semantic analyzing can be true and the searching efficiency will increase. The node document is categorized by semantic, and the node groups between each other founds relationship by semantic similarity. The transmission of serching request is realized by the friendship in the serching process. The simulating results indicate the SBSGSS mechanism can highly improve the serching efficiency.