基于多蚁型的蚁群聚类算法

Ant Colony Clustering Algorithm Based on Multiple Heterogeneous Ants

本文提出了一种多蚁型的蚁群模型和在此模型上的一个具有移动空间的蚁群聚类算法。每个类簇都带有一个移动空间,使同类的蚂蚁紧密地聚在一起。移动空间为每个类簇选出一个代表蚂蚁用来长距离的类簇移动。代表蚂蚁通过连接将类簇中不相似的大量数据移动到与其相似的代表蚂蚁处,减少了蚂蚁的搜索时间,提高聚类性能。针对UCI数据集进行实验,结果表明新算法提高了聚类效果。

The paper presents an ant colony model based on multiple heterogeneous ants and an ant colony clustering algorithm with movement zones. Here ants maintain a movement zone around each cluster, keeping ants close to data items. Each movement zone elect the representatives that are responsible for all long distance movement. The representatives can, probabilistically, pass a large number of dissimilar objects it has to any other similar representative, greatly reducing cluster merging time and improving the performance of clustering. Experimental results show that the new algorithm can improve the quality of clustering.