DocumentCode
921686
Title
On the application of Bashkirov, Braverman, and Muchnik potential function for feature selection in pattern recognition
Author
Babu, C.C. ; Kalra, S.N.
Volume
60
Issue
3
fYear
1972
fDate
3/1/1972 12:00:00 AM
Firstpage
333
Lastpage
334
Abstract
Based on the potential function suggested by Bashkirov, Braverman, and Muchnik for pattern classification, a criterion is proposed for the selection of effective pattern features. Furthermore, its relationship with the upperbound estimate of the probability of misrecognition is presented.
Keywords
Bayesian methods; Councils; Covariance matrix; Density functional theory; Euclidean distance; Pattern analysis; Pattern classification; Pattern recognition; Smoothing methods;
fLanguage
English
Journal_Title
Proceedings of the IEEE
Publisher
ieee
ISSN
0018-9219
Type
jour
DOI
10.1109/PROC.1972.8631
Filename
1450561
Link To Document