Aggregation invariance in general clustering approaches

Details

Ressource 1Download: BIB_E83647B5E790.P001.pdf (242.33 [Ko])
State: Public
Version: Author's accepted manuscript
Serval ID
serval:BIB_E83647B5E790
Type
Article: article from journal or magazin.
Collection
Publications
Institution
Title
Aggregation invariance in general clustering approaches
Journal
Advances in Data Analysis and Classification
Author(s)
Bavaud F.
ISSN
1862-5347 (Print1862-5355)
Publication state
Published
Issued date
11/2009
Peer-reviewed
Oui
Volume
3
Number
3
Pages
205-225
Language
english
Abstract
General clustering deals with weighted objects and fuzzy memberships. We investigate the group- or object-aggregation-invariance properties possessed by the relevant functionals (effective number of groups or objects, centroids, dispersion, mutual object-group information, etc.). The classical squared Euclidean case can be generalized to non-Euclidean distances, as well as to non-linear transformations of the memberships, yielding the c-means clustering algorithm as well as two presumably new procedures, the convex and pairwise convex clustering. Cluster stability and aggregation-invariance of the optimal memberships associated to the various clustering schemes are examined as well.
Keywords
Aggregation invariance - c-Means clustering - EM algorithm - Fuzzy clustering - Model-based clustering - Mutual information
Open Access
Yes
Create date
22/12/2009 14:44
Last modification date
20/08/2019 16:11
Usage data