Cluste-Based Social Search Engine Design And Implementation

The rapid development of network, Internet users through the network to obtain information. Traditional search engines in resource acquisition has limitations, the current build on the search engine’s social search engine to enhance user query information coverage and the accuracy of search results have become an active research area in the field. At present numerous social search engine to search results and user retrieval used to classification or in tissue, the retrieval results still adopted a linear list, complex process, users find their required information results quite difficult, so, if you want to improve information retrieval speed as well as the information needed for the rapid positioning, the retrieval result clustering processing becomes the most effective solution.This study combines the association rules and the method of cluster analysis, in view of social search engine technology and the application of data mining technology, in the realization of a clustering of social search engine system based on SCMES, the realization of the system process to make the detailed in-depth analysis.The use of search engine parameters database, to achieve the following several aspects of the content, the realization of a retrieval request, the conversion of the search engine inquiry instruction; secondly, to search results page to related information extraction; on the results recorded in the title and abstract word processing to extract key words, establish associated word matrix, Boolean function and the TFIDF function to express the result of feature vectors, the use of K-means clustering scheme to test and analysis to the performance of the system, according to the application example of the corresponding analysis comparison.