Computer Engineering and Applications ›› 2009, Vol. 45 ›› Issue (11): 145-146.DOI: 10.3778/j.issn.1002-8331.2009.11.044

• 数据库、信号与信息处理 • Previous Articles     Next Articles

Improved k-nearest neighbor classification algorithm

SANG Ying-bin,LIU Qiong-sun   

  1. College of Mathematics and Physics,Chongqing University,Chongqing 400044,China
  • Received:2008-02-26 Revised:2008-05-06 Online:2009-04-11 Published:2009-04-11
  • Contact: SANG Ying-bin

改进的k-nn快速分类算法

桑应宾,刘琼荪   

  1. 重庆大学 数理学院,重庆400044
  • 通讯作者: 桑应宾

Abstract: In order to overcome the disadvantages of traditional k-nn,this paper uses two algorithms of classification and clustering to proposes an improved k-nn classification algorithm.Experiments show that this algorithm can speed up when it has a few effects in accuracy.

摘要: 针对传统的k-近邻(k-nn)方法的缺点,将聚类中的K均值和分类中的k近邻算法有机结合,提出了一种改进的k-nn快速分类算法。实验表明该算法在影响分类效果不大的情况下能达到快速分类的目的。