DocumentCode :
918200
Title :
Error probability in dependent pattern classification (Corresp.)
Author :
Forsyth, John J.
Volume :
18
Issue :
5
fYear :
1972
fDate :
9/1/1972 12:00:00 AM
Firstpage :
678
Lastpage :
680
Abstract :
This correspondence derives an upper bound on the probability of error of the 
-class Bayes decision process when the patterns observed have first-order stochastic dependence. The bound is derived by applying an information-theoretic approach in which both the equivocation and the Bhattacharyya coefficient play a role.
-class Bayes decision process when the patterns observed have first-order stochastic dependence. The bound is derived by applying an information-theoretic approach in which both the equivocation and the Bhattacharyya coefficient play a role.Keywords :
Bayes procedures; Pattern classification; Computer science; Entropy; Error probability; Machine learning; Mutual information; Pattern classification; Pattern recognition; Random variables; Stochastic processes; Upper bound;
fLanguage :
English
Journal_Title :
Information Theory, IEEE Transactions on
Publisher :
ieee
ISSN :
0018-9448
Type :
jour
DOI :
10.1109/TIT.1972.1054883
Filename :
1054883
Link To Document :
