nav emailalert searchbtn searchbox tablepage yinyongbenwen piczone journalimg journalInfo journalinfonormal searchdiv searchzone qikanlogo popupnotification paper paperNew
2010, 03, v.42 59-62
一种改进的蚁群聚类算法
基金项目(Foundation):
邮箱(Email):
DOI:
移动端阅读
摘要:

针对现有蚁群聚类中将带聚类样本放于网格进行聚类的算法存在随机移动而延长聚类时间,及大数据集进行蚁群聚类时收敛速度慢的缺点,在蚁群进行聚类前增加数据预处理.利用两元素越相似属于同一类簇的可能性越大的思想,将样本集中的样本量缩小.研究了通过信息素进行聚类的蚁群聚类算法,使算法中的"蚂蚁"在一定指导下进行聚类,达到缩短时间的目的.最后通过实验验证了所提出算法的有效性和优越性.

Abstract:

To shorten clustering time in ant colony algorithm(ACA) and speed up convergence rate of large data sets,data preprocessing is adopted before ant colony clustering algorithm(ACCA).Meanwhile,clustering speed is studied through the pheromone of ACCA,and ants in the algorithm should be guided by certain information.In order to test the validity of the algorithms,K-means and the basic ant colony clustering are compared at the same time.The experimental results show the effectiveness of the proposed approach.

基本信息:

中图分类号:TP181

引用信息:

[1]俞辉,裴振奎,陈继东.一种改进的蚁群聚类算法[J],2010,42(03):59-62.

检 索 高级检索