Title :
A criterion based on an information theoretic measure for goodness of fit between classifier and data base
Author :
Eigen, D.J. ; Davida, G.I. ; Northouse, R.A.
Author_Institution :
Bell Laboratories
Abstract :
A criterion for characterizing an iteratively trained classifier is presented. The criterion is based on an information theoretic measure that is developed from modeling classifier training iterations as a set of cascaded channels. The criterion is formulated as a figure of merit and as a performance index to check the appropriateness of application of the characterized classifier to anunknown data base and for implementing classifier updates and data selection respectively.
Conference_Titel :
Decision and Control including the 12th Symposium on Adaptive Processes, 1973 IEEE Conference on
Conference_Location :
San Diego, CA, USA
DOI :
10.1109/CDC.1973.269111
