Methods for Designing Multiple Classifier Systems
In the field of pattern recognition, multiple classifier systems based on the combination of outputs of a set of different classifiers have been proposed as a method for the development of high performance classification systems. In this paper, the problem of design of multiple classifier system is discussed. Six design methods based on the so-called “overproduce and choose“ paradigm are described and compared by experiments. Although these design methods exhibited some interesting features, they do not guarantee to design the optimal multiple classifier system for the classification task at hand. Accordingly, the main conclusion of this paper is that the problem of the optimal MCS design still remains open.
Unable to display preview. Download preview PDF.
- Kittler J., Roli F. (eds.): Multiple Classifier Systems. Proc. of the First Int. Workshop MCS 2000. Lecture Notes in Computer Science, Vol. 1857. Springer-Verlag (2000)Google Scholar
- Kittler J.: A framework for classifier fusion: is it still needed? In: Ferri F.J., Inesta J.M., Amin A., Pudil P. (eds.): Advances in Pattern Recognition. Proc. Int. Workshop SSPR&SPR 2000. Lectures Notes in Computer Science, Vol. 1876. Springer-Verlag (2000) 45–56Google Scholar
- Ho T.K.: Complexity of classification problems and comparative advantages of combined classifiers. In: Kittler J., Roli F. (eds.): Multiple Classifier Systems. Proc. of the First Int. Workshop MCS 2000. LNCS Vol. 1857 Springer-Verlag (2000) 97–106Google Scholar
- Giacinto G., Roli F.: Design of effective neural network ensembles for image classification purposes. to appear in Image and Vision Computing Journal (2001)Google Scholar