Please use this identifier to cite or link to this item:
Title: Order invariant hierarchical clustering
Authors: Lee, JWT
Yeung, DS
Hui-Chan, C
Tam, SF
Keywords: Hierarchical clustering
Order invariance
Issue Date: 2002
Source: Proceedings of the IEEE International Conference on Systems, Man and Cybernetics, 2002, v. 2, p. 497-502 How to cite?
Abstract: Clustering is an important area in data analysis in which we try to find natural groupings of objects based on their similarity (or alternatively, dissimilarity). In many studies, information on dissimilarity are basically qualitative. In such situations an order invariant approach to cluster analysis is desirable. Furthermore, in areas of study such as biological taxonomy or cognitive science, we often look for hierarchical clustering of objects where object groupings can be examined at different levels of refinement. Traditionally order invariant approaches to hierarchical clustering are based on some form of graph connectivity, or measures such as cluster diameter and separation. In this paper, we propose a new approach to order invariant hierarchical clustering based on an ordinal consistency perspective that is suitable for qualitative or subjective dissimilarity data. We show some experimental results that demonstrate the advantages of our method.
Description: 2002 IEEE International Conference on Systems, Man and Cybernetics, Yasmine Hammamet, 6-9 October 2002
ISSN: 0884-3627
Appears in Collections:Conference Paper

View full-text via PolyU eLinks SFX Query
Show full item record


Last Week
Last month
Citations as of Jun 1, 2018

Page view(s)

Last Week
Last month
Citations as of Jun 17, 2018

Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.