Title :
Performance considerations for a combined information classification test using Dirichlet priors
Author :
Lynch, Robert S. ; Willett, Jr Peter K
Author_Institution :
Naval Undersea Warfare Center, Newport, RI, USA
fDate :
6/1/1999 12:00:00 AM
Abstract :
The average probability of error is used to demonstrate performance of the combined Bayes test (CBT), which assumes a uniform Dirichlet prior for the symbol probabilities of each class. The performance of the CBT is compared with an uncombined maximum likelihood (ML) based test and with the Kolmogorov-Smirnov test
Keywords :
Bayes methods; error statistics; pattern classification; Dirichlet priors; Kolmogorov-Smirnov test; average error probability; combined Bayes test; combined information classification test; performance; symbol probabilities; uncombined maximum likelihood based test; uniform Dirichlet prior; Bayesian methods; Encoding; Laboratories; Performance evaluation; Probability; Quantization; Statistics; Testing; Training data;
Journal_Title :
Signal Processing, IEEE Transactions on
