一种新的粗糙Leader聚类算法

New Rough Leader Clustering Algorithm

聚类是数据挖掘领域重要的研究方向。在众多的聚类算法中,Leader算法运用很广泛,但Leader算法没有考虑到聚类分析中内在的不确定性。对Leader算法做了相应改进,加入了粗糙集和粒计算的思想,使其能够处理聚类中固有的不确定性,得到更合理的聚类结果。最后,通过实验证明了该算法的优越性。

Clustering is a major research orientation in data mining. Among all the clustering algorithms, leader algorithm is widely used, but it fails to take into consideration the inherent uncertainty involved in clustering analysis. This paper proposed an improved leader algorithm based on rough Set and granular computing. The novel leader algorithm can deal with the intrinsic uncertainty in clustering analysis and make the clustering results more reasonable. Finally, the superiority of the new rough Leader algorithm is proved by experimentation.