Skip to Main content Skip to Navigation
Journal articles

Mesures de distance entre modalités de variables qualitatives; application à la classification

Hisham Abdallah 1 Gilbert Saporta 2
2 CEDRIC - MSDMA - CEDRIC. Méthodes statistiques de data-mining et apprentissage
CEDRIC - Centre d'études et de recherche en informatique et communications
Abstract : Clustering categories needs a definition of dissimilarities between modalities. Our treatment will be based on Ochiai, Dice etc. indices of similarity. We give bounds and check axioms of distance for some indices of dissimilarity. We then use them to get agglomerative hierarchical clustering of a set of modalities.
Complete list of metadatas

Cited literature [21 references]  Display  Hide  Download

https://hal-cnam.archives-ouvertes.fr/hal-02507690
Contributor : Gilbert Saporta <>
Submitted on : Monday, March 16, 2020 - 2:53:26 PM
Last modification on : Thursday, March 19, 2020 - 11:42:35 AM

File

RSA_2003__51_2_75_0.pdf
Publisher files allowed on an open archive

Identifiers

  • HAL Id : hal-02507690, version 1

Collections

Citation

Hisham Abdallah, Gilbert Saporta. Mesures de distance entre modalités de variables qualitatives; application à la classification. Revue de Statistique Appliquée, Société française de statistique, 2003, 51 (2), pp.75-90. ⟨hal-02507690⟩

Share

Metrics

Record views

27

Files downloads

6