Abstract
This paper is basically concerned with a practical problem: the on-the-fly quick learning of handwritten character recognition systems. More generally, it explores the problem of generating new learning examples, especially from very scarce (2 to 5 per class) original learning data. It presents two different methods. The first one is based on applying distortions on original characters using knowledge on handwriting properties like speed, curvature etc. The second one consists in generation based on the notion of analogical dissimilarity which quantifies the analogical relation “A is to B almost as C is to D”. We give an algorithm to compute the k-least dissimilar objects D, hence generating k new objects from three examples A, B and C. Finally, we experimentally prove the efficiency of both methods, especially when used in conjunction.
Chapter PDF
Similar content being viewed by others
Keywords
References
Wolf, L., Martin, I.: Robust boosting for learning from few examples. In: IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), IEEE Computer Society Press, Los Alamitos (2005)
Bishop, C.: Training with noise is equivalent to Tikhonov regularization. Neural Computation 7(1), 108–116 (1995)
Cano, J., Pérez-Cortes, J.C., Arlandis, J., Llobet, R.: Training set expansion in handwritten character recognition. In: Proc. of the 9th Int. Workshop on Structural and Syntactic Pattern Recognition, pp. 548–556 (2002)
Simard, P., Steinkraus, D., Platt, J.C.: Best practice for convolutional neural network applied to visual analysis. In: Proc. of the 7th Int. Conf. on Document Analysis and Recognition (2003)
Varga, T., Bunke, H.: Generation of synthetic data for an HMM-based handwriting recognition system. In: Proc. of 7th Int. Conf. on Document Analysis and Recognition, pp. 618–622 (2003)
Bishop, C.: Pattern Recognition and Machine Learning. Springer, Heidelberg (2007)
Anquetil, E., Lorette, G.: Perceptual model of handwriting drawing application to the handwriting segmentation problem. In: Proc. of the 4th Int. Conf. on Document Analysis and Recognition, pp. 112–117 (1997)
Hesse, M.: Aristotle’s logic of analogy. The Philosophical Quarterly 15(61), 328–340 (1965)
Gentner, D., Holyoak, K.J., Kokinov, B.: The analogical mind: Perspectives from cognitive science. MIT Press, Cambridge (2001)
Lepage, Y.: Solving analogies on words: an algorithm. In: Proc. of COLING-ACL 1998, vol. 1, pp. 728–735 (1998)
Bayoudh, S., Miclet, L., Delhay, A.: Learning by analogy: a classification rule for binary and nominal data. In: Proc. of the Int. Joint Conf. on Artificial Intelligence, vol. 20, pp. 678–683 (2007)
Nilsson, N.: Principles of Artificial Intelligence. Tioga Publishing Company (1980)
Eppstein, D.: Finding the k shortest paths. SIAM J. Computing 28(2), 652–673 (1998)
Varga, T., Kilchhofer, D., Bunke, H.: Template-based synthetic handwriting generation for the training of recognition systems. In: Proc. of 12th Conf. of the International Graphonomics Society, pp. 206–211 (2005)
Plamondon, R., Guerfali, W.: The generation of handwriting with delta-lognormal synergies. Biological Cybernetics 78, 119–132 (1998)
Gillick, L., Cox, S.J.: Some statistical issues in the comparison of speech recognition algorithms. In: IEEE (ed.) Proceedings of IEEE International Conference on Acoustics, Speech, and Signal Processing, Glasgow, Scotland, pp. 532–535. IEEE Computer Society Press, Los Alamitos (1989)
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 2007 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Bayoudh, S., Mouchère, H., Miclet, L., Anquetil, E. (2007). Learning a Classifier with Very Few Examples: Analogy Based and Knowledge Based Generation of New Examples for Character Recognition. In: Kok, J.N., Koronacki, J., Mantaras, R.L.d., Matwin, S., Mladenič, D., Skowron, A. (eds) Machine Learning: ECML 2007. ECML 2007. Lecture Notes in Computer Science(), vol 4701. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-74958-5_49
Download citation
DOI: https://doi.org/10.1007/978-3-540-74958-5_49
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-74957-8
Online ISBN: 978-3-540-74958-5
eBook Packages: Computer ScienceComputer Science (R0)