Based On The Node Of Interest For Unstructured P2P Network Search Mechanism

With the development of economy, the growing popularity of the Internet,network resources and the number of Internet users was explosive growth, willinevitably lead to the emergence of a problem, it is such a huge ocean resources, howfast an accurate and comprehensive targeted to what you want resources? Face moreand more resources, the traditional disadvantages of server-based search engine moreand more obvious, such as by the impact of computing power and storage capacity ofthe server, greatly affect the search efficiency, P2P technology as a new networkorganizational model for the development of the search engine space search algorithmproposed, to help improve search efficiency.The lack of global topological structure compared to structured P2P networks,unstructured P2P network node, the random dynamic changes more complex, how toefficiently search the resources of interest to the user is the key problems of theunstructured P2P network search strategy. The most extensive search strategy inunstructured P2P networks using the flooding mechanism redundant messages, searchthe exponential growth, but with the massive number of network nodes and thenumber of resources, seriously affect the user’s search efficiency.Solve this problem, we propose an unstructured P2P search strategy based onuser interest, have the resources well to express interest in the node, the node theyshowed interest in the network, so that the degree of interest high node in the networkis relatively close. When the users in the network to find the resources of interest,only the query message is forwarded to the node has resources, thereby increasingsearch efficiency.The main content of this article:(1) First of all, the concept and role of the P2P network are described, and thenmade a detailed introduction to the characteristics of structured and unstructured P2P network and search algorithm, these are the basics of this thesis;(2) This paper introduces the based on HowNet words similarity calculationdetails HowNet word similarity calculation process, word similarity calculation is thebasis of the proposed algorithm;(3) The proposed collection of high-frequency k-vector to express the interest ofthe nodes, the method is proposed on the basis of the k-vector of high-frequencywords, better able to express interest in the node;(4) Node degree of interest in the P2P network-based search algorithm, cosinealgorithm to calculate the node of interest related to interest high node associated withthe nodes with the same resources in the same region of interest, to search for suchresources, just to the corresponding domain of interest to find, which greatly improvesthe search efficiency. With the increase in the number of searches, the node neighbornode table optimization, making the search for increasingly efficient;(5) Finally, the proposed algorithm simulation experiments and existing node ofinterest-based search algorithm comparison shows that: this algorithm greatly reducethe leak investigation, and improves the overall rate of the Find and search successrate.