Article (Scientific journals)
Classification performance resulting from of 2-means
Ruwet, Christel; Haesbroeck, Gentiane
2013In Journal of Statistical Planning and Inference, 143 (2), p. 408-418
Peer Reviewed verified by ORBi
 

Files


Full Text
Classification Performance resulting from a 2-means.pdf
Author preprint (325.27 kB)
Download
Full Text Parts
Classification Performance resulting from a 2-means_R3.pdf
Author postprint (244.7 kB)
Request a copy

All documents in ORBi are protected by a user license.

Send to



Details



Keywords :
Asymptotic loss; Cluster analysis; Error rate; k-means; Influence function; Principal points; Robustness
Abstract :
[en] The k-means procedure is probably one of the most common nonhierachical clustering techniques. From a theoretical point of view, it is related to the search for the k principal points of the underlying distribution. In this paper, the classification resulting from that procedure for k=2 is shown to be optimal under a balanced mixture of two spherically symmetric and homoscedastic distributions. Then, the classification efficiency of the 2-means rule is assessed using the second order influence function and compared to the classification efficiencies of the Fisher and logistic discriminations. Influence functions are also considered here to compare the robustness to infinitesimal contamination of the 2-means method w.r.t. the generalized 2-means technique.
Disciplines :
Mathematics
Author, co-author :
Ruwet, Christel ;  Université de Liège - ULiège > Département de mathématique > Statistique mathématique
Haesbroeck, Gentiane ;  Université de Liège - ULiège > Département de mathématique > Statistique mathématique
Language :
English
Title :
Classification performance resulting from of 2-means
Publication date :
February 2013
Journal title :
Journal of Statistical Planning and Inference
ISSN :
0378-3758
Publisher :
Elsevier Science
Volume :
143
Issue :
2
Pages :
408-418
Peer reviewed :
Peer Reviewed verified by ORBi
Available on ORBi :
since 09 January 2011

Statistics


Number of views
129 (21 by ULiège)
Number of downloads
101 (5 by ULiège)

Scopus citations®
 
4
Scopus citations®
without self-citations
4
OpenCitations
 
5

Bibliography


Similar publications



Contact ORBi