COMPARISON OF PATTERN-RECOGNITION METHODS BY PRECISION OF APPROXIMATING THE SEPARATING HYPERPLANES

Authors
BOLOTOV AA LARICHEV O
Citation
Aa. Bolotov et O. Larichev, COMPARISON OF PATTERN-RECOGNITION METHODS BY PRECISION OF APPROXIMATING THE SEPARATING HYPERPLANES, Automation and remote control, 56(7), 1995, pp. 1004-1010
Citations number
11
Categorie Soggetti
Controlo Theory & Cybernetics","Computer Application, Chemistry & Engineering","Instument & Instrumentation","Robotics & Automatic Control
Journal title
Automation and remote control → ACNP
ISSN journal
00051179
Volume
56
Issue
7
Year of publication
1995
Part
2
Pages
1004 - 1010
Database
ISI
SICI code
0005-1179(1995)56:7<1004:COPMBP>2.0.ZU;2-Y
Abstract
A new criterion, The precision of approximating The separating hyperpl anes. as proposed for comparison of the pattern recognition methods. U sing statistical modeling, two methods of pattern recognition - the pr obabilistic method based on the Bayes formula and that of a linear dis criminant function - are compared for four types of boundaries between the decision classes.