On the identity of optimal strategies for multistage classifiers
暂无分享,去创建一个
Abstract This paper deals with the decision rules of a multistage classifier based on a decision tree scheme. Two optimal (Bayes) strategies for performing the classification at each nonterminal node are derived. The first type optimal strategy uses features connected with particular nodes of a tree, whereas the second type optimal strategy takes into account all the measured features. Some conditions are given for which both of these strategies are identical.
[1] M. Kurzynski,et al. Diagnosis of acute abdominal pain using a three-stage classifier. , 1987, Computers in biology and medicine.
[2] Marek Kurzynski,et al. On the multistage Bayes classifier , 1988, Pattern Recognit..