Class-confidence critic combining

M. Aksela1,2, R. Girdziugas3, J. Laaksonen1, E. Oja1, J. Kangas4
1Helsinki University of Technology, Laboratory of Computer and Information Science, Finland
2Aalto-yliopisto Insinooritieteiden korkeakoulu, Aalto, FI
3Lab. of Comput. & Inf. Sci., Helsinki Univ. of Technol., Finland
4Nokia Research Center, Tampere, Finland

Tóm tắt

This paper discusses a combination of two techniques for improving the recognition accuracy of on-line handwritten character recognition: committee classification and adaptation to the user. A novel adaptive committee structure, namely the class-confidence critic combination (CCCC) scheme, is presented and evaluated. It is shown to be able to improve significantly on its member classifiers. Also the effect of having either more or less diverse sets of member classifiers is considered.

Từ khóa

#Conferences #Handwriting recognition

Tài liệu tham khảo

10.1109/34.368145 girdziusas, 2001, Discriminative On-line Recognition of Isolated Handwritten Character 10.1142/S0218001493000352 10.1007/PL00013555 schwaighofer, 2001, SVM Toolbox for Matlab (Version v0 4 russell, 1995, Artifi Cial Intelligence A Modern Approach 10.1109/78.790663 xiao, 2000, Adapti ve combination of classi-ers and its application to handwritten chinese character recognition, Proceedings of the 15th International Confer-ence on P Attern Reco Gnition, 2, 327, 10.1109/ICPR.2000.906079