Email this article Multilevel models for solution of multiclass recognition problems
- Authors: Dokukin A.A.1, Ryazanov V.V.2, Shut O.V.3
-
Affiliations:
- Dorodnicyn Computing Centre, Federal Research Center Computer Science and Control
- Moscow Institute of Physics and Technology
- Belarusian State University
- Issue: Vol 26, No 3 (2016)
- Pages: 461-473
- Section: Methematical Method in Pattern Recognition
- URL: https://bakhtiniada.ru/1054-6618/article/view/194784
- DOI: https://doi.org/10.1134/S1054661816030044
- ID: 194784
Cite item
Open Access
Access granted
Subscription Access
Abstract
The problem of choosing an ensemble of binary subproblems for multiclass recognition problems is considered in the algebraic and logical approaches. The connection of the first and second level algorithm correctness and the applicability of logical deduction in a multilevel scheme are studied. Two modifications of the ECOC method are proposed that allow one to improve the quality of the original method under different conditions; this is demonstrated using model and applied problems.
About the authors
A. A. Dokukin
Dorodnicyn Computing Centre, Federal Research Center Computer Science and Control
Author for correspondence.
Email: dalex@ccas.ru
Russian Federation, ul. Vavilova 44, Moscow, 119333
V. V. Ryazanov
Moscow Institute of Physics and Technology
Email: dalex@ccas.ru
Russian Federation, Institutskii per. 9, Dolgoprudnyi, Moscow oblast, 141700
O. V. Shut
Belarusian State University
Email: dalex@ccas.ru
Belarus, pr. Nezavisimosti 4, Minsk, 220030
Supplementary files
