基于最近邻聚类的INTERNET信息检索系统

随着网络的不断普及和发展，Internet为用户提供了一个极有价值的信息源。如何能够快速、准确的检索出用户感兴趣的信息，已经成为当前研究的热点。本文分析了KNN算法和聚类方法，指出了其中的不足并在此基础上提出了一种改进算法。该算法将聚类方法和KNN算法的优点结合起来，从而达到缩减了训练样本数量，减少了算法计算量，加快检索速度的目的。

Information Retrieval System for Internet based on Nearest Neighbor-clustering

Internet is more and more widely used, which provide a valuable information resource for users. How retrieve the information users prefere rapidly and accurately become the focus nowadays. K nearest neighbor and clustering being analysized, point out demerits in existence, and based on these present a improving algorithm. This algorithm combine advantages of KNN and Clustering, decreasing training samples and quantity of algorithm calculating, and increasing the speed of retrieval.