计算机工程与应用 ›› 2009, Vol. 45 ›› Issue (9): 138-140.DOI: 10.3778/j.issn.1002-8331.2009.09.040

• 数据库、信号与信息处理 • 上一篇    下一篇

基于属性约简的聚类分析技术

彭 云,丁树良   

  1. 江西师范大学 计算机信息工程学院,南昌 330022
  • 收稿日期:2008-02-01 修回日期:2008-04-09 出版日期:2009-03-21 发布日期:2009-03-21
  • 通讯作者: 彭 云

Clustering analysis based on attributes reduction

PENG Yun,DING Shu-liang   

  1. College of Computer Information Engineering,Jiangxi Normal University,Nanchang 330022,China
  • Received:2008-02-01 Revised:2008-04-09 Online:2009-03-21 Published:2009-03-21
  • Contact: PENG Yun

摘要: 由于原始海量数据规模较大,聚类算法难以实现,而且聚类分析有时候只考虑关键属性作为分类参数。因此提出了一种对原始数据先进行属性约简,再聚类分析的策略,可以在一定程度上消除数据对象中的冗余属性,改善聚类分析的鲁棒性,提高聚类算法的有效性。

Abstract: The scale of original data is very large.It is difficult to realize the clustering algorithm.Clustering analysis often takes the key attributes as classification parameters.This paper presents a method of clustering analysis based on attributes reduction.It can reduce the redundant attributes,and improve the robusting and efficiency of clustering analysis.