Unpublished conference/Abstract (Scientific congresses and symposiums)
Classification efficiency of the trimmed k-means procedure
Ruwet, Christel
201244e Journées de Statistique
 

Files


Full Text
PresentationJdS2012.pdf
Publisher postprint (1.8 MB)
Request a copy
Annexes
RésuméLongJDS.pdf
Publisher postprint (80.94 kB)
Download

All documents in ORBi are protected by a user license.

Send to



Details



Keywords :
Classification; Efficiency; Influence function; Optimality; Robustness
Abstract :
[en] The k-means method is used in classification to group similar observations in k groups. When a second sample is available to test the obtained groupings, the rate of misclassification can be computed. If the samples are generated from a mixture of two homoscedastic and spherically symmetric distributions, the rate of misclassification equals that of the Bayes rule. Therefore, the k-means method is optimal under such a mixture model. However, it is not robust with respect to outliers in the dataset used to construct the groups. To avoid this problem, the k-means procedure has been adapted in many ways. This presentation focuses on the trimmed k-means method defined by trimming some of the observations. The advantage of this method, besides its resistance to outliers, is that optimality is preserved. However, it is well known that trimming observations leads to a loss in classification efficiency. The latter can be measured by means of the influence function of the misclassifiation rate.
Disciplines :
Mathematics
Author, co-author :
Ruwet, Christel ;  Université de Liège - ULiège > Département de mathématique > Statistique mathématique
Language :
English
Title :
Classification efficiency of the trimmed k-means procedure
Alternative titles :
[fr] Efficacité de classification de la méthode des k-moyennes tronquées
Publication date :
21 May 2012
Event name :
44e Journées de Statistique
Event organizer :
Société Française de Statistique
Event place :
Bruxelles, Belgium
Event date :
21-25 mai 2012
Audience :
International
Available on ORBi :
since 12 August 2013

Statistics


Number of views
53 (7 by ULiège)
Number of downloads
469 (5 by ULiège)

Bibliography


Similar publications



Contact ORBi