On evidential clustering with partial supervision - Archive ouverte HAL Access content directly
Book Sections Year : 2018

On evidential clustering with partial supervision

Abstract

This paper introduces a new semi-supervised evidential clustering algorithm. It considers label constraints and exploits the evidence theory to create a credal partition coherent with the background knowledge. The main characteristics of the new method is its ability to express the uncertainties of partial prior information by assigning each constrained object to a set of labels. It enriches previous existing algorithm that allows the preservation of the uncertainty in the constraint by adding the possibility to favor crisp decision following the inherent structure of the dataset. The advantages of the proposed approach are illustrated using both a synthetic dataset and a real genomics dataset.
Fichier principal
Vignette du fichier
antoine2018evidential.pdf (211.72 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-02787897 , version 1 (26-01-2023)

Identifiers

Cite

Violaine Antoine, Kevin Gravouil, Nicolas Labroche. On evidential clustering with partial supervision. Belief Functions: Theory and Applications, Chap. 3, Editions Springer, 2018, Lecture Notes in Computer Science, 978-3-319-99382-9. ⟨10.1007/978-3-319-99383-6_3⟩. ⟨hal-02787897⟩
31 View
0 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More