Selection of Classifiers Based on Multiple Classifier Behaviour

In the field of pattern recognition, the concept of Multiple Classifier Systems (MCSs) was proposed as a method for the development of high performance classification systems. At present, the common “operation” mechanism of MCSs is the “combination” of classifiers outputs. Recently, some researchers pointed out the potentialities of “dynamic classifier selection” (DCS) as a new operation mechanism. In this paper, a DCS algorithm based on the MCS behaviour is presented. The proposed method is aimed to exploit the behaviour of the MCS in order to select, for each test pattern, the classifier that is more likely to provide the correct classification. Reported results on the classification of different data sets show that dynamic classifier selection based on MCS behaviour is an effective operation mechanism for MCSs.