Doctoral thesis (Dissertations and theses)
Robustness analysis of clustering and classification techniques
Ruwet, Christel
2012
 

Files


Full Text
These.pdf
Author preprint (2.58 MB)
Request a copy
Annexes
DefenseBis.pdf
Publisher postprint (1.07 MB)
Slides de la défense
Request a copy
Erratum.pdf
Publisher postprint (32.01 kB)
Erratum
Request a copy

All documents in ORBi are protected by a user license.

Send to



Details



Keywords :
Clustering; Classification; Error rate; Breakdown point; Influence function; Robustness
Abstract :
[en] As mentioned in the title, the framework of this doctoral dissertation encompasses two different subjects: robust statistics on the one hand and classification and clustering techniques on the other hand. Robust procedures try at the same time to emulate classical procedures and to produce results that are not unduly affected by contaminated observations or deviations from model assumptions. Classification and clustering techniques try to find groups among observations. Grouping is one of the most basic abilities of living creatures; the simple fact of naming objects is already grouping. The main interest lies in the fact that the characteristics of a group as well as its differences from other groups can be used as a summary of the dataset.
Disciplines :
Mathematics
Author, co-author :
Ruwet, Christel ;  Université de Liège - ULiège > Département de mathématique > Statistique mathématique
Language :
English
Title :
Robustness analysis of clustering and classification techniques
Defense date :
08 June 2012
Institution :
ULiège - Université de Liège
Degree :
Docteur en Sciences
Promotor :
Haesbroeck, Gentiane ;  Université de Liège - ULiège > Mathematics
Garcia-Escudero, Luis Angel
President :
Albert, Adelin  ;  Université de Liège - ULiège > Département des sciences de la santé publique
Jury member :
Heuchenne, Cédric ;  Université de Liège - ULiège > HEC Recherche > HEC Recherche: Business Analytics & Supply Chain Management
Croux, Christophe
Dehon, Catherine
Gordaliza, Alfonso
Available on ORBi :
since 06 May 2012

Statistics


Number of views
153 (47 by ULiège)
Number of downloads
19 (18 by ULiège)

Bibliography


Similar publications



Contact ORBi