Abstract
This paper proposes a new method for discovering piecewise polynomials to fit multivariate data containing numeric and nominal variables. Each polynomial is accompanied with the corresponding nominal condition stating a subspace where the polynomial is applied. Such a nominally conditioned polynomial is called a rule. A set of such rules can be represented as a single numeric function, which can be approximated by a four-layer perceptron. The method selects the best from those trained perceptrons, and finds the final rules from the best perceptron. In our experiments the proposed method finds the original rules for an artificial data set, and discovers succinct rules for a real data set.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Falkenhainer, B.C., Michalski, R.S.: Integrating quantitative and qualitative discovery in the ABACUS system. In: Machine Learning: An Artificial Intelligence Approach, vol. 3, pp. 153–190. Morgan Kaufmann, San Francisco (1990)
Lloyd, S.P.: Least squares quantization in PCM. IEEE Trans. Information Theory IT-28(2), 129–137 (1982)
Nakano, R., Saito, K.: Discovery of a set of nominally conditioned polynomials. In: Arikawa, S., Furukawa, K. (eds.) DS 1999. LNCS (LNAI), vol. 1721, pp. 287–298. Springer, Heidelberg (1999)
Nakano, R., Saito, K.: Finding polynomials to fit multivariate data having numeric and nominal variables. In: Hoffmann, F., Adams, N., Fisher, D., Guimarães, G., Hand, D.J. (eds.) IDA 2001. LNCS, vol. 2189, pp. 258–267. Springer, Heidelberg (2001)
Nakano, R., Saito, K.: Discovering polynomials to fit multivariate data having numeric and nominal variables. In: Arikawa, S., Shinohara, A. (eds.) Progress in Discovery Science. LNCS (LNAI), vol. 2281, pp. 482–493. Springer, Heidelberg (2002)
Quinlan, J.R.: C4.5: programs for machine learning. Morgan Kaufmann, San Francisco (1993)
Saito, K., Nakano, R.: Partial BFGS update and efficient step-length calculation for three-layer neural networks. Neural Computation 9(1), 239–257 (1997)
Saito, K., Nakano, R.: Discovery of a set of nominally conditioned polynomials using neural networks, vector quantizers, and decision trees. In: Morishita, S., Arikawa, S. (eds.) DS 2000. LNCS (LNAI), vol. 1967, pp. 325–329. Springer, Heidelberg (2000)
Stone, M.: Cross-validatory choice and assessment of statistical predictions (with discussion). Journal of the Royal Statistical Society B 64, 111–147 (1974)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2004 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Tanahashi, Y., Saito, K., Nakano, R. (2004). Piecewise Multivariate Polynomials Using a Four-Layer Perceptron. In: Negoita, M.G., Howlett, R.J., Jain, L.C. (eds) Knowledge-Based Intelligent Information and Engineering Systems. KES 2004. Lecture Notes in Computer Science(), vol 3214. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-30133-2_79
Download citation
DOI: https://doi.org/10.1007/978-3-540-30133-2_79
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-23206-3
Online ISBN: 978-3-540-30133-2
eBook Packages: Springer Book Archive