期刊名称：International Journal of Computer Science Issues
印刷版ISSN：1694-0784
电子版ISSN：1694-0814
出版年度：2014
卷号：11
期号：2
出版社：IJCSI Press
摘要：Cluster ensemble algorithms have been used in different fields like data mining, bioinformatics and pattern recognition. Many of them use label correspondence as a step which can be performed with some accuracy if all the input partitions are generated with same k. Thus these algorithms produce good results if this k is close to the actual number of clusters in the dataset. This puts great restriction if user has no idea of the number of clusters. In this paper we show through experimental studies that good ensembles can be generated even if the input solutions contain different number of clusters.
