Computer Engineering and Applications ›› 2008, Vol. 44 ›› Issue (23): 154-156.DOI: 10.3778/j.issn.1002-8331.2008.23.047

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

Outlier mining based on swarm intelligence

ZHANG Ran1,2,JIA Rui-yu1,QIAN Guang-chao1   

  1. 1.School of Computer Science and Technology,Anhui University,Hefei 230039,China
    2.Department of Computer,Tongling College,Tongling,Anhui 244000,China
  • Received:2007-10-16 Revised:2007-12-28 Online:2008-08-11 Published:2008-08-11
  • Contact: ZHANG Ran

基于群体智能的离群数据挖掘

张 然1,2,贾瑞玉1,钱光超1   

  1. 1.安徽大学 计算机科学与技术学院,合肥 230039
    2.铜陵学院 计算机系,安徽 铜陵 244000
  • 通讯作者: 张 然

Abstract: Outlier mining is an important task of data mining.This paper firstly analyzes outlier and outlier mining algorithms.Then according to LF algorithm and CSI algorithm,an outlier mining algorithm based on swarm intelligence is proposed,and an experiment is conducted.Results show that the validity of outlier mining algorithm based on swarm intelligence.Compared with the other algorithm,the algorithm in this paper has better robustness because it avoids the influence when users initialize the value of parameters,and do not need set original clustering centers.

Key words: outlier mining, swarm intelligence, clustering

摘要: 离群数据挖掘是数据挖掘的重要任务之一。首先分析了离群数据及其挖掘方法,然后根据LF算法和CSI算法,提出了基于群体智能的离群数据挖掘算法,并进行了仿真实验。实验结果显示了基于群体智能的离群数据挖掘算法的有效性。与其它方法相比,该算法避免了用户在设定参数初始值时给算法带来的影响,并且不需要设定初始聚类中心,因此具有更好的鲁棒性。

关键词: 离群数据挖掘, 群体智能, 聚类