en
Scientific article
Open access
English

Information–Theoretic Multiclass Classification Based on Binary Classifiers

Published inJournal of Signal Processing Systems, vol. 65, no. 3, p. 413-430
Collection
  • Open Access - Licence nationale Springer 
Publication date2011
Abstract

In this paper, we consider the multiclass classification problem based on sets of independent binary classifiers. Each binary classifier represents the output of a quantized projection of training data onto a randomly generated orthonormal basis vector thus producing a binary label. The ensemble of all binary labels forms an analogue of a coding matrix. The properties of such kind of matrices and their impact on the maximum number of uniquely distinguishable classes are analyzed in this paper from an information-theoretic point of view. We also consider a concept of reliability for such kind of coding matrix generation that can be an alternative to other adaptive training techniques and investigate the impact on the bit error probability. We demonstrate that it is equivalent to the considered random coding matrix without any bit reliability information in terms of recognition rate.

Citation (ISO format)
VOLOSHYNOVSKYY, Svyatoslav et al. Information–Theoretic Multiclass Classification Based on Binary Classifiers. In: Journal of Signal Processing Systems, 2011, vol. 65, n° 3, p. 413–430. doi: 10.1007/s11265-010-0526-1
Main files (1)
Article (Published version)
accessLevelPublic
Identifiers
ISSN of the journal1939-8018
444views
212downloads

Technical informations

Creation03/13/2019 2:48:00 PM
First validation03/13/2019 2:48:00 PM
Update time03/15/2023 3:59:48 PM
Status update03/15/2023 3:59:47 PM
Last indexation01/17/2024 5:06:49 AM
All rights reserved by Archive ouverte UNIGE and the University of GenevaunigeBlack