On a Classification Method for a Large Number of Classes
- Авторы: Zhuravlev Y.I.1, Ryazanov V.V.2, Aslanyan L.H.3, Sahakyan H.A.3
-
Учреждения:
- Dorodnitsyn Computing Center, Federal Research Center Informatics and Control, Russian Academy of Sciences
- Moscow Institute of Physics and Technology (State University)
- Institute for Informatics and Automation Problems, National Academy of Sciences of Armenia
- Выпуск: Том 29, № 3 (2019)
- Страницы: 366-376
- Раздел: Mathematical Method in Pattern Recognition
- URL: https://journals.rcsi.science/1054-6618/article/view/195617
- DOI: https://doi.org/10.1134/S1054661819030246
- ID: 195617
Цитировать
Аннотация
The construction of a two-level decision scheme for recognition problems with many classes is proposed that is based on the development of the error-correcting output codes (ЕСОС) method. In the “classical” ЕСОС, a large number of partitions of the original classes into two macroclasses are constructed. Each macroclass is a union of some original classes. Each macroclass is assigned either 0 or 1. As a result, each original class is defined by a row of 0 and 1 (the stage of encoding) and a coding matrix is constructed. The stage of classification of an arbitrary new object consists in the solution of each dichotomic problem and application of a special decision rule (the stage of decoding). In this paper, new methods for weighting and taking into account codewords, modifying decision rules, and searching for locally optimal dichotomies are proposed, and various quality criteria for classification and the cases of extension of a codeword are considered.
Об авторах
Yu. Zhuravlev
Dorodnitsyn Computing Center, Federal Research Center Informatics and Control,Russian Academy of Sciences
Email: hsahakyan@sci.am
Россия, Moscow, 119333
V. Ryazanov
Moscow Institute of Physics and Technology (State University)
Автор, ответственный за переписку.
Email: vasily.ryazanov@phystech.edu
Россия, Dolgoprudny, Moscow oblast, 141700
L. Aslanyan
Institute for Informatics and Automation Problems, National Academy of Sciences of Armenia
Email: hsahakyan@sci.am
Армения, Yerevan
H. Sahakyan
Institute for Informatics and Automation Problems, National Academy of Sciences of Armenia
Автор, ответственный за переписку.
Email: hsahakyan@sci.am
Армения, Yerevan
Дополнительные файлы
