Dhillon, I. S.; Modha, D. S. & Spangler, W. S.
(2002):
Class visualization of highdimensional data with applications.
In: Computational Statistics & Data Analysis,
Ausgabe/Number: 1,
Vol. 41,
Erscheinungsjahr/Year: 2002.
Seiten/Pages: 5990.
[Volltext] [Kurzfassung] [BibTeX]
[Endnote]
No abstract is available for this item.
@article{RePEc:eee:csdana:v:41:y:2002:i:1:p:5990,
author = {Dhillon, Inderjit S. and Modha, Dharmendra S. and Spangler, W. Scott},
title = {Class visualization of highdimensional data with applications},
journal = {Computational Statistics & Data Analysis},
year = {2002},
volume = {41},
number = {1},
pages = {5990},
url = {http://www.cs.utexas.edu/~inderjit/public_papers/csda.pdf},
keywords = {alphaworks, cluster, clustering, dm, visualization},
abstract = {No abstract is available for this item.}
}
%0 = article
%A = Dhillon, Inderjit S. and Modha, Dharmendra S. and Spangler, W. Scott
%D = 2002
%T = Class visualization of highdimensional data with applications
%U = http://www.cs.utexas.edu/~inderjit/public_papers/csda.pdf
Bezdek, J. C.; Li, W. Q.; Attikiouzel, Y. & Windham, M.
(1997):
A geometric approach to cluster validity for normal mixtures.
In: Soft Computing  A Fusion of Foundations, Methodologies and Applications,
Ausgabe/Number: 4,
Vol. 1,
Erscheinungsjahr/Year: 1997.
Seiten/Pages: 166179.
[Volltext] [Kurzfassung] [BibTeX]
[Endnote]
We study indices for choosing the correct number of components in a mixture of normal distributions. Previous studies have been confined to indices based wholly on probabilistic models. Viewing mixture decomposition as probabilistic clustering (where the emphasis is on partitioning for geometric substructure) as opposed to parametric estimation enables us to introduce both fuzzy and crisp measures of cluster validity for this problem. We presume the underlying samples to be unlabeled, and use the expectationmaximization (EM) algorithm to find clusters in the data. We test 16 probabilistic, 3 fuzzy and 4 crisp indices on 12 data sets that are samples from bivariate normal mixtures having either 3 or 6 components. Over three run averages based on different initializations of EM, 10 of the 23 indices tested for choosing the right number of mixture components were correct in at least 9 of the 12 trials. Among these were the fuzzy index of XieBeni, the crisp DaviesBouldin index, and two crisp indices that are recent generalizations of Dunn's index.

@article{bezdek1997,
author = {Bezdek, J. C. and Li, W. Q. and Attikiouzel, Y. and Windham, M.},
title = {A geometric approach to cluster validity for normal mixtures},
journal = {Soft Computing  A Fusion of Foundations, Methodologies and Applications},
year = {1997},
volume = {1},
number = {4},
pages = {166179},
url = {http://dx.doi.org/10.1007/s005000050019},
keywords = {cluster, evaluation, index},
abstract = {We study indices for choosing the correct number of components in a mixture of normal distributions. Previous studies have been confined to indices based wholly on probabilistic models. Viewing mixture decomposition as probabilistic clustering (where the emphasis is on partitioning for geometric substructure) as opposed to parametric estimation enables us to introduce both fuzzy and crisp measures of cluster validity for this problem. We presume the underlying samples to be unlabeled, and use the expectationmaximization (EM) algorithm to find clusters in the data. We test 16 probabilistic, 3 fuzzy and 4 crisp indices on 12 data sets that are samples from bivariate normal mixtures having either 3 or 6 components. Over three run averages based on different initializations of EM, 10 of the 23 indices tested for choosing the right number of mixture components were correct in at least 9 of the 12 trials. Among these were the fuzzy index of XieBeni, the crisp DaviesBouldin index, and two crisp indices that are recent generalizations of Dunn's index.
ER }
}
%0 = article
%A = Bezdek, J. C. and Li, W. Q. and Attikiouzel, Y. and Windham, M.
%D = 1997
%T = A geometric approach to cluster validity for normal mixtures
%U = http://dx.doi.org/10.1007/s005000050019
Kaufman, L. & Rousseeuw, P. J. (Hrsg.)
(1990):
Finding Groups in Data: An Introduction to Cluster Analysis.
Erscheinungsjahr/Year: 1990.
Verlag/Publisher: John Wiley,
[BibTeX]
[Endnote]
@book{kaufman1990finding,
author = {Kaufman, L. and Rousseeuw, P. J.},
title = {Finding Groups in Data: An Introduction to Cluster Analysis},
publisher = {John Wiley},
year = {1990},
isbn = {1581331097},
keywords = {clustering, evaluation, cluster}
}
%0 = book
%A = Kaufman, L. and Rousseeuw, P. J.
%D = 1990
%I = John Wiley
%T = Finding Groups in Data: An Introduction to Cluster Analysis
Rand, W.
(1971):
Objective criteria for the evaluation of clustering methods.
In: Journal of the American Statistical Association ,
Ausgabe/Number: 336,
Vol. 66,
Erscheinungsjahr/Year: 1971.
Seiten/Pages: 846850.
[BibTeX]
[Endnote]
@article{rand1971,
author = {Rand, W.M.},
title = {Objective criteria for the evaluation of clustering methods},
journal = {Journal of the American Statistical Association },
year = {1971},
volume = {66},
number = {336},
pages = {846850},
keywords = {cluster, clustering, criteria, evaluation, index, rand}
}
%0 = article
%A = Rand, W.M.
%D = 1971
%T = Objective criteria for the evaluation of clustering methods