期刊名称：International Journal of Computer Science and Information Technologies
电子版ISSN：0975-9646
出版年度：2015
卷号：6
期号：3
页码：2620-2622
出版社：TechScience Publications
摘要：The number of people in the Web available, it is relevant to their interests involved people finding people has become more difficult for users increases. According to some of the often closely-each subset data (ideally) some common traits that stock so subsets of clustering, classification of a data set (cluster) at defined measuring distance. It is easier to find relevant people and also provided for social network applications that understand the various aspects of the query form to help users can enable users to. Clustering algorithm is based on a popular technique for n has been divided into groups such that. In this method, called the cluster centers groups that are identified by a set of points. The data points that is closest to the center of the cluster. Clustering the existing algorithms is slow and the large number of data points and initializations vary depending on local minima to converge. A fast clustering algorithm can attack both these shortcomings, but the large number of data points for this algorithm is used when there is a limit, then we to calculate the distortion algorithm to introduce an effective way. Experiment results fast algorithm is better than other methods and compared on the basis of relevance ranking users more easily find the relevant ones can help.
关键词：Peoples clustering; fast greedy k-means;Search engine.
