Skip to Main content Skip to Navigation
Poster communications

Détermination du nombre des classes dans l’algorithmeCROKI2 de classification croisée

Abstract : One of the major problems of clustering is how to identify the number of clusters. Thisproblem is also present in biclustering or block clustering methods. In this paper, we areinterested by CROKI2 algorithm proposed by Govaert (1983) for contingency tables. Our goalis to find the optimal number of clusters on rows and columns by using some techniques ofcluster validity.
Document type :
Poster communications
Complete list of metadata

https://hal.archives-ouvertes.fr/hal-01125739
Contributor : Laboratoire Cedric <>
Submitted on : Wednesday, December 16, 2020 - 12:43:05 PM
Last modification on : Thursday, January 14, 2021 - 12:29:23 PM
Long-term archiving on: : Wednesday, March 17, 2021 - 7:10:56 PM

File

EGC 2009 version finale.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01125739, version 1

Collections

Citation

Malika Charrad, Yves Lechevallier, Gilbert Saporta, Mohamed Ben Ahmed. Détermination du nombre des classes dans l’algorithmeCROKI2 de classification croisée. EGC'2009: 9ème conférence internationale sur l'Extraction et la Gestion des Connaissances, Jan 2009, Strasbourg, France. pp.447-448, 2009. ⟨hal-01125739⟩

Share

Metrics

Record views

138

Files downloads

11