Skip to Main content Skip to Navigation
Conference papers

Arbres de classification construits à partir de fonctions de croyance

Nicolas Sutton-Charani 1 Sébastien Destercke 2 Thierry Denoeux 3
1 LGI2P - Laboratoire de Génie Informatique et Ingénierie de Production
2 Labex MS2T - Laboratoire d'Excellence "Maîtrise des Systèmes de Systèmes Technologiques"
3 Heudiasyc - Heuristique et Diagnostic des Systèmes Complexes [Compiègne]
Abstract : Decision trees are popular classification methods. In paper, we extend to multi-class problems a decision tree method based on belief functions previously described for two-class problems only. We propose three possible extensions : combining multiple two-class trees together and directly extending the estimation of belief functions within the tree to the multi-class setting. We provide experiments, results and compare them to usual decision trees.
Keywords : decision trees belief function Classification
Document type :
Conference papers
Complete list of metadatas

Cited literature [18 references]  Display  Hide  Download
https://hal.archives-ouvertes.fr/hal-01923310
Contributor : Nicolas Sutton-Charani <>
Submitted on : Thursday, November 15, 2018 - 10:48:25 AM
Last modification on : Wednesday, June 24, 2020 - 4:18:15 PM
Long-term archiving on: : Saturday, February 16, 2019 - 1:08:44 PM

File

Vignette du document
N. Sutton-Charani - LFA 2012.p...
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01923310, version 1

Collections

CNRS | EM-ALES | UNIV-COMPIEGNE | MS2T | HEUDIASYC | DI | INSTITUT-TELECOM

Citation

Nicolas Sutton-Charani, Sébastien Destercke, Thierry Denoeux. Arbres de classification construits à partir de fonctions de croyance. Rencontres francophones sur la Logique Floue et ses Applications (LFA 2012), Nov 2012, Compiègne, France. ⟨hal-01923310⟩

Export

BibTeX TEI DC DCterms EndNote

Share

Metrics

Record views

94

Files downloads

38