In the field of pattern recognition, the concept of multiple classifier systems (MCS) was proposed as a method for the development of high-performance classification systems. At present, the common "operation" mechanism of MCS is the "combination" of classifier outputs. Recently, some researchers have pointed out the potentialities of "dynamic classifier selection" as a new operation mechanism. In a previous paper, the authors discussed the advantages of "selection-based" MCS and proposed an algorithm for dynamic classifier selection. In this paper, a theoretical framework for dynamic classifier selection is described and two methods for selecting classifiers are proposed. Reported results on the classification of different data sets show that dynamic classifier selection is an effective method for the development of MCS.
Methods for dynamic classifier selection
1999-01-01
52720 byte
Conference paper
Electronic Resource
English
Methods for Dynamic Classifier Selection
British Library Conference Proceedings | 1999
|British Library Online Contents | 2015
|Comparison of Genetic Algorithm and Sequential Search Methods for Classifier Subset Selection
British Library Conference Proceedings | 2003
|Network Traffic Classification Using Dynamic State Classifier
British Library Conference Proceedings | 2005
|