International Journal of Computer Science Issues
Hierarchical clustering is often depicted as the better quality clustering approach, but is limited because of its second degree complexity. The distributed clustering area aims to solve several problems that currently limit the scalability of network resources. While, clustering methods determine relationships among the objects, and allow the determination of similar groups of objects. This paper is to partition the network into such a set of clusters, which have observed similar phenomena. This paper presents the results of an experimental study of some comparisons clustering techniques: K-means, and hybrid hierarchical-K-means.