A New Simplified Gravitational Clustering Method for Multi-prototype Learning Based on Minimum Classification Error Training

In this paper, we propose a new simplified gravitational clustering method for multi-prototype learning based on minimum classification error (MCE) training. It simulates the process of the attraction and merging of objects due to their gravity force. The procedure is simplified by not considering velocity and multi-force attraction. The proposed hierarchical method does not depend on random initialization and the results can be used as better initial centers for K-means to achieve higher performance under the SSE (sum-squared-error) criterion. The experimental results on the recognition of handwritten Chinese characters show that the proposed approach can generate better prototypes than K-means and the results obtained by MCE training can be further improved when the proposed method is employed.

[1]  Ahmad Fuad Rezaur Rahman,et al.  Multi-prototype classification: improved modelling of the variability of handwritten data using statistical clustering algorithms , 1997 .

[2]  Biing-Hwang Juang,et al.  Pattern recognition using a family of design algorithms based upon the generalized probabilistic descent method , 1998, Proc. IEEE.

[3]  D.M. Mount,et al.  An Efficient k-Means Clustering Algorithm: Analysis and Implementation , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[4]  Biing-Hwang Juang,et al.  Discriminative learning for minimum error classification [pattern recognition] , 1992, IEEE Trans. Signal Process..

[5]  Qiang Huo,et al.  High performance Chinese OCR based on Gabor features, discriminative feature extraction and model training , 2001, 2001 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings (Cat. No.01CH37221).

[6]  Teuvo Kohonen,et al.  The self-organizing map , 1990, Neurocomputing.

[7]  Pavel Berkhin,et al.  A Survey of Clustering Data Mining Techniques , 2006, Grouping Multidimensional Data.

[8]  Yen-Jen Oyang,et al.  A statistics-based approach to control the quality of subclusters in incremental gravitational clustering , 2005, Pattern Recognit..

[9]  Masaki Nakagawa,et al.  'Online recognition of Chinese characters: the state-of-the-art , 2004, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[10]  Rongchun Zhao,et al.  Match between normalization schemes and feature sets for handwritten Chinese character recognition , 2001, Proceedings of Sixth International Conference on Document Analysis and Recognition.