Abstract
To model a complex manufacturing process effectively with commonly used machine learning methods (Rule Induction, MLP, ANFIS) further preprocessing steps of feature selection, feature value characterization and feature value smoothing are required. The model prediction error serves as the fitness measure. It depends strongly on the parameter settings of the processing modules. Qualitatively better processing parameter settings are found by iterative training process where these settings are modified based on the quality of the previous learning process, controlled by downhill algorithms (Evolutionary Strategy, Downhill Simplex). This enables us to find a suitable model describing a process with less effort. The traditional optimizing process of determining signal processing parameters with heuristics may be standardized and preserved through this mechanism.
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
Buckley, J.J., Hayashi, Y.: Fuzzy neural networks: A survey. Fuzzy sets and systems 66, 1–13 (1994)
Dietterich, T.G.: Machine Learning Research - Four current Directions. AI Magazine, 108 (Winter 1997)
Jang, J.-S.R.: ANFIS: Adaptive Network-based Fuzzy Inference System. IEEE Transactions on Systems, Man and Cybernetics 23(3) (May 1993)
McGill, W.: Multivariate Information Transmission. IEEE Transactions on Information Theory 1, 93–111 (1955)
Nelder, J.A., Mead, R.: Computer Journal. 7, 308–313 (1965)
Press, W., et al.: Downhill Simplex Method in Multidimensions, Numerical Recipes in C. Cambridge University Press, Cambridge (1996)
Quinlan, J.R.: nduction of Decision Trees. Machine Learning 1, 81–106 (1986); C4.5: Programs for Machine Learning. Morgan Kaufmann, San Francisco (1993)
Rechenberg, I.: Optimierung technischer Systeme nach Prinzipien der biologischen Evolution. Frommann Holzboog (1973)
Rosenblatt, F.: The perceptron: a probabilistic model for information storage and Organization in the Brain. Psychological Preview 65, 386–408 (1988)
Rummelhart, D.E., McClelland, J.L.: Parallel distributed Processing, ch.8, pp. 318–362. MIT Press, Cambridge (1986)
Shannon, C.E.: A mathematical Theory of Communication. Bell Systems Technology Journal 27, 379–423 (1948)
Weiss, S.M., Indurkhya, N.: Reduced Complexity Rule Induction. Learn and Knowledge Acquisition, 678–684 (1997)
Wettschereck, D., Aha, D.W., Mohri, T.: Technical Report AIC-95-012, Navel Research Laboratory, Navy Center for Applied Research in Artificial Intelligence, Washington, D.C. (1997)
Zadeh, L.: Fuzzy Sets, Information and Control (1965)
Zell, A., et al.: SNNS User Manual 4.0 (1995), ftp://ftp.informatik.uni-stuttgart.de
Zimmermann, H.-J.: Fuzzy Set Theory and its Applications. Kluwer-Nijhoff Publishing, Boston (1988)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1999 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Link, M., Ishitobi, M. (1999). Supervised Parameter Optimization of a Modular Machine Learning System. In: Imam, I., Kodratoff, Y., El-Dessouki, A., Ali, M. (eds) Multiple Approaches to Intelligent Systems. IEA/AIE 1999. Lecture Notes in Computer Science(), vol 1611. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-48765-4_67
Download citation
DOI: https://doi.org/10.1007/978-3-540-48765-4_67
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-66076-7
Online ISBN: 978-3-540-48765-4
eBook Packages: Springer Book Archive