Previous |  Up |  Next


training statistical pattern; similarity measures
In this paper the possibilities are discussed for training statistical pattern recognizers based on a distance representation of the objects instead of a feature representation. Distances or similarities are used between the unknown objects to be classified with a selected subset of the training objects (the support objects). These distances are combined into linear or nonlinear classifiers. In this approach the feature definition problem is replaced by finding good similarity measures. The proposal corresponds with determining classification functions in Hilbert space using an infinite feature set. It is a direct consequence of Vapnik’s support vector classifier [Vap2].
[1] Aizerman M. A., Braverman E. M., Rozonoer L. I.: The probability problem of pattern recognition learning and the method of potential functions. Automat. Remote Control 25 (1964), 1175-1193 MR 0172768
[2] Devijver P. A., Kittler J.: Pattern Recognition: A Statistical Approach. Prentice Hall, London 1982 MR 0692767 | Zbl 0542.68071
[3] Duin R. P. W.: Small sample size generalization. In: SCIA’95, Proc. 9th Scandinavian Conf. on Image Analysis (G. Borgefors, ed.), Volume 2, Uppsala 1995, pp. 957–964
[4] Duin R. P. W., Ridder D. de: Neural network experiences between perceptrons and support vectors. In: Proc. of the 8th British Machine Vision Conference (A. F. Clark, ed.), Volume 2, Colchester 1997, pp. 590–599
[5] Duin R. P. W., Ridder D. de, Tax D. M. J.: Experiments with object based discriminant functions; a featureless approach to pattern recognition. In: Pattern Recognition in Practice V, Vlieland, 1997, to be published in Pattern Recognition Letters
[6] Duin R. P. W., Ridder D. de, Tax D. M. J.: Featureless Classification. In: Proc. 1st International Workshop Statistical Techniques in Pattern Recognition (P. Pudil, J. Novovičová and J. Grim, eds.), Prague 1997, pp. 37–42
[7] Jain A. K., Chandrasekaran B.: Dimensionality and sample size considerations in pattern Recognition practice. In: Handbook of Statistics (P. R. Krishnaiah and L. N. Kanal, eds.), Vol. 2, North–Holland, Amsterdam 1987, pp. 835–855
[8] Raudys S.: Evolution and generalization of a single neurone. I. Single layer perceptron as seven statistical classifiers. Neural Networks, to be published
[9] Schölkopf B.: Support Vector Learning. Ph.D. Thesis, Techn. Universität Berlin 1997 Zbl 0935.68084
[10] Tax D. M. J., Ridder D. de, Duin R. P. W.: Support vector classifiers: a first look. In: ASCI’97, Proc. Third Annual Conference of the Advanced School for Computing and Imaging, 1997
[11] Vapnik V. N.: Estimation of Dependences Based on Empirical Data. Springer–Verlag, New York 1982 MR 0672244 | Zbl 1118.62002
[12] Vapnik V. N.: The Nature of Statistical Learning Theory. Springer–Verlag, Berlin 1995 MR 1367965 | Zbl 0934.62009
[13] Wilson C. L., Marris M. D.: Handprinted Character Database 2. National Institute of Standards and Technology; Advanced Systems division, 1990
Partner of
EuDML logo