Abstract
Many machine learning methods just consider the quality of prediction results as their final purpose. To make the prediction process transparent (reversible), spline kernel based methods were proposed by Gunn. However, the original solution method, termed SUpport vector Parsimonious ANOVA (SUPANOVA) was computationally very complex and demanding. In this paper, we propose a new heuristic to compute the optimal sparse vector in SUPANOVA that replaces the original solver for the convex quadratic problem of very high dimensionality. The resulting system is much faster without the loss of precision, as demonstrated in this paper on two benchmarks: the iris data set and the Boston housing market data benchmark.
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
Vapnik, V.: The Nature of Statistical Learning Theory. Springer, New York (1995)
Wold, S., Sjöström, M., Eriksson, L.: PLS-Regression: a Basic Tool of Chemometrics. Chemometrics and Intelligent Laboratory Systems 58, 109–130 (2001)
Gunn, S., Brown, M.: Supanova - a sparse, transparent modelling approach. In: Proc. IEEE Workshop on Neural Networks for Signal Processing, Madison, WI, pp. 21–30. IEEE Computer Society Press, Los Alamitos (1999)
Gunn, S., Kandola, J.: Structural Modelling with Sparse Kernels. Machine Learning 48(1), 137–163 (2002)
Embrechts, M., et al.: Use of Machine Learning for Classification of Magnetocardiograms. In: Proc. IEEE Conference on System, Man and Cybernetics Washington DC, pp. 1400–1405. IEEE Computer Society Press, Los Alamitos (2003)
Embrechts, M., Szymanski, B., Sternickel, K.: Introduction to Scientific Data Mining. In: Computationally Intelligent Hybrid Systems: The Fusion of Soft Computing and Hard Computing, Wiley, New York (2005)
Kandola, J.: Structural Modelling with Sparse Kernels. Department of Electronics and Computer Science, University of Southampton, U.K. (2001)
Meszaros, C.: The BPMPD interior point solver for convex quadratic problems. TR WP 98-8, Computer and Automation Institute, Academy of Sciences, Budapest, Hungary (1998)
Merz, C.J., Murphy, P.M.: UCI repository for machine learning data-bases (1996), http://www.ics.uci.edu/~mlearn/MLRepository.html
Fisher, A.: The use of multiple measurements in taxonomic problems. Annual Eugenics 7, 179–188 (1936)
Harrison, D., Rubinfield, D.: Hedonic housing prices and the demand for clean air. Journal of Enviromental Economics and Management 5, 81–102 (1978)
Neal, R.: Bayesian Learning for Neural Networks. Springer, New York (1995)
Husmeier, D.: Neural Networks for Conditional Probability Estimation. Springer, New York (1999)
Szymanski, B., et al.: Using efficient supanova kernel for heart disease diagnosis. In: Proc. 16th Conf. Artificial Neural Networks Industrial Engineering, ANNIE (2006)
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 2007 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Szymanski, B.K., Zhu, L., Han, L., Embrechts, M., Ross, A., Sternickel, K. (2007). A Computationally Efficient SUPANOVA: Spline Kernel Based Machine Learning Tool. In: Saad, A., Dahal, K., Sarfraz, M., Roy, R. (eds) Soft Computing in Industrial Applications. Advances in Soft Computing, vol 39. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-70706-6_14
Download citation
DOI: https://doi.org/10.1007/978-3-540-70706-6_14
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-70704-2
Online ISBN: 978-3-540-70706-6
eBook Packages: EngineeringEngineering (R0)