Some Properties of Binary Classifier with Fuzzy-Valued Loss Function
In this paper we present some prosperities of binary classifier with fuzzyvalued loss function. The loss function in our case is dependent on the stage of the decision tree or depends on the node of the decision tree. The decision rules of a two-stage binary classifier minimize the mean risk, that is the mean value of the fuzzy loss function. In the paper the effect of a loss function on the value of the separation point of decision regions is presented. In this paper we will are not going to study the impact of the choice of ranking of fuzzy numbers method on the results of the classification.
KeywordsDecision Rule Loss Function Fuzzy Number Recognition Algorithm Recognition Problem
Unable to display preview. Download preview PDF.