You need JavaScript to view this

Recursive automatic classification algorithms

Journal Article:

Abstract

A variational statement of the automatic classification problem is given. The dependence of the form of the optimal partition surface on the form of the classification objective functional is investigated. A recursive algorithm is proposed for maximising a functional of reasonably general form. The convergence problem is analysed in connection with the proposed algorithm. 8 references.
Publication Date:
Mar 01, 1982
Product Type:
Journal Article
Reference Number:
EDB-85-152435
Resource Relation:
Journal Name: Avtom. Telemekh.; (USSR); Journal Volume: 3
Subject:
99 GENERAL AND MISCELLANEOUS//MATHEMATICS, COMPUTING, AND INFORMATION SCIENCE; CLASSIFICATION; ALGORITHMS; ARTIFICIAL INTELLIGENCE; AUTOMATION; INFORMATION THEORY; OPTIMIZATION; VARIATIONAL METHODS; MATHEMATICAL LOGIC; 990200* - Mathematics & Computers
OSTI ID:
5362115
Country of Origin:
USSR
Language:
English
Other Identifying Numbers:
Journal ID: CODEN: AVTEA
Submitting Site:
HEDB
Size:
Pages: 95-105
Announcement Date:

Journal Article:

Citation Formats

Bauman, E V, and Dorofeyuk, A A. Recursive automatic classification algorithms. USSR: N. p., 1982. Web.
Bauman, E V, & Dorofeyuk, A A. Recursive automatic classification algorithms. USSR.
Bauman, E V, and Dorofeyuk, A A. 1982. "Recursive automatic classification algorithms." USSR.
@misc{etde_5362115,
title = {Recursive automatic classification algorithms}
author = {Bauman, E V, and Dorofeyuk, A A}
abstractNote = {A variational statement of the automatic classification problem is given. The dependence of the form of the optimal partition surface on the form of the classification objective functional is investigated. A recursive algorithm is proposed for maximising a functional of reasonably general form. The convergence problem is analysed in connection with the proposed algorithm. 8 references.}
journal = {Avtom. Telemekh.; (USSR)}
volume = {3}
journal type = {AC}
place = {USSR}
year = {1982}
month = {Mar}
}