摘要
提出了一种基于蚂蚁孵化分类行为的聚类算法:预先定好簇数,把数据随机放入簇中,通过蚂蚁把混乱无章的簇整理成有序.该算法与传统蚂蚁聚类算法相比,模型更直观,操作更简单.对Iris等真实数据的实验结果证明,该算法快速而且有效.
A hatchery-based ant clustering algorithm was explored.With the pre-configured number of clusters,the data were randomly placed into the clusters,then arranged the initial disorder clusters orderly through ants.Compared with other traditional ant clustering algorithms,this algorithm is a more intuitive model,and easier operation.The experiments with real data as Iris showed that the algorithm is fast and effective.
出处
《郑州轻工业学院学报（自然科学版）》
CAS
2009年第5期64-67,共4页
Journal of Zhengzhou University of Light Industry:Natural Science
基金
河南省自然基金项目(082300410280)
河南省教育厅自然基金项目(2006520019)
关键词
数据挖掘
聚类分析
蚂蚁算法
data mining
clustering analysis
ant algorithm
