计算机工程与应用 ›› 2009, Vol. 45 ›› Issue (29): 131-133.DOI: 10.3778/j.issn.1002-8331.2009.29.039

• 数据库、信号与信息处理 • 上一篇    下一篇

基于K-均值聚类和凝聚聚类的离群点查找方法

曾 颖,罗 可,邹瑞芝   

  1. 长沙理工大学 计算机与通信工程学院,长沙 410076
  • 收稿日期:2008-06-02 修回日期:2009-04-22 出版日期:2009-10-11 发布日期:2009-10-11
  • 通讯作者: 曾 颖

Outliers detection method based on K-means and agglomerative clustering

ZENG Ying,LUO Ke,ZOU Rui-zhi   

  1. College of Computer and Communication Engineering,Changsha University of Science and Technology,Changsha 410076,China
  • Received:2008-06-02 Revised:2009-04-22 Online:2009-10-11 Published:2009-10-11
  • Contact: ZENG Ying

摘要: 离群点发现是数据挖掘研究的一个重要方面。根据数据流的特点,给出了一种基于K-均值聚类和凝聚聚类的离群点发现方法,先用K-均值聚类对数据流进行处理,生成中间聚类结果,然后用凝聚聚类对这些中间结果进行再次选择,最后找出可能存在的离群点。

关键词: 数据挖掘, 离群点, K-均值聚类, 凝聚聚类

Abstract: Outliers detection is an important issue in data mining.In this paper,according to the characteristics of data streams,an outliers detection method based on k-means and agglomerative clustering is proposed,which uses k-means clustering to find some intermediate results,and applies agglomerative clustering on each intermediate result to pick out potential outliers.

Key words: data mining, outliers, k-means clustering, agglomerative clustering

中图分类号: