计算机工程与应用 ›› 2008, Vol. 44 ›› Issue (16): 154-156.

• 数据库、信号与信息处理 • 上一篇    下一篇

一种基于混合策略的蚁群聚类算法

郭会林,苏一丹   

  1. 广西大学 计算机与电子信息学院,南宁 530004
  • 收稿日期:2007-09-12 修回日期:2007-12-07 出版日期:2008-06-01 发布日期:2008-06-01
  • 通讯作者: 郭会林

New ant clustering algorithm based on multi-strategies

GUO Hui-lin,SU Yi-dan   

  1. School of Computer,Electronics and Information,Guangxi University,Nanning 530004,China
  • Received:2007-09-12 Revised:2007-12-07 Online:2008-06-01 Published:2008-06-01
  • Contact: GUO Hui-lin

摘要: 蚁群算法是一种智能聚类分析方法。分析了基本蚁群聚类算法的特点和不足,总结了不同的聚类情景,并在此基础上提出了一种新的基于混合策略的蚁群聚类算法。新算法中,蚂蚁根据不同的聚类情景而采取不同的行为策略,同时赋予蚂蚁多载功能。实验表明基于混合策略的蚁群聚类算法显著改善了聚类效果。

关键词: 蚁群聚类算法, 混合策略, 多载

Abstract: Ant colony algorithm is an intelligent clustering analysis methodology.This paper analyzes the feedback of the basic ant clustering algorithm,concludes the clustering scenes,and presents a new ant clustering algorithm based on multi-strategies.In the new algorithm,ants act differently according to the clustering scenes and they can multi-load.The experiment shows the new algorithm performs better than Acluster.

Key words: ant clustering algorithm, multi-strategies, multi-load