期刊
IEEE TRANSACTIONS ON FUZZY SYSTEMS
卷 18, 期 1, 页码 27-39出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TFUZZ.2009.2034531
关键词
Clustering combination; dendrogram descriptor; fuzzy-equivalence relation; hierarchical clustering; min-transitive closure; ultrametric property
资金
- Iran Telecommunication Research Center [T/500/20615]
Clustering-combination methods have received considerable attentions in recent years, and many ensemble-based clustering methods have been introduced. However, clustering-combination techniques have been limited to flat clustering combination, and the combination of hierarchical clusterings has yet to be addressed. In this paper, we address and formalize the concept of hierarchical-clustering combination and introduce an algorithmic framework in which multiple hierarchical clusterings could be easily combined. In this framework, the similarity-based description matrices of input hierarchical clusterings are aggregated into a transitive consensus matrix in which the final hierarchy could be formed. Empirical evaluation, by using popular available datasets, confirms the superiority of combined hierarchical clustering introduced by our method over the standard (single) hierarchical-clustering methods.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据