Abstract
Knowledge acquisition is a frequent bottleneck in artificial intelligence applications. Neural learning may offer a new perspective in this field. Using Self-Organising Neural Networks, as the Kohonen model, the inherent structures in high-dimensional input spaces are projected on a low dimensional space. The exploration of structures resp. classes is then possible applying the U-Matrix method for the visualisation of data. Since Neural Networks are not able to explain the obtained results, a machine learning algorithm sig* was developed to extract symbolic knowledge in form of rules out of subsymbolic data. Combining both approaches in hybrid system results in a powerful method to solve classification and diagnosis problems. Several applications have been used to test this method. Applications on processes with dynamic characteristics, such as chemical processes and avalanche forecasting show that an extension of this method from static to dynamic data is feasible.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Deichsel, G,; Trampisch, H.J.: Clusteranalyse und Diskriminanzanlyse. Stuttgart: G. Fischer Verlag, 1985.
Föhn, P.; Hächler, P.: Prevision des grosses avalanches au moyens d’un model deterministic-statistique. ANENA. in: Comples Rendues du 2e Rencontre Int. sur la Neige et des Avalanches, Grenoble, 12–14, 1978, pp.151–156.
Guimaraes, G. & Korus, D. „Neuronale Netze auf Transputern“Transputer-Anwender-Treffen TAT’92, Abstraktband, Aachen Sep. 1992.
Kohonen, Teuvo „Self-Organization and Associative Memory“Springer Verlag 1989 (3rd ed.)
Müller, F.; Seifert, O.: Taschenbuch der medizinisch-klinischen Diagnostik, Berlin: Springer-Verlag, 72 Aufl., 1989.
Quinlan, J.R.: Learning Efficient Classification Procedures and their Application to Chess End Games, in: Michalsky, R.; Carbonell, J.G; Mitchell, T.M.(Hrsg.): Machine Learning - An Artificial Intelligence Approach. Berlin 1984, pp. 463–482.
Schweizer, M.; Föhn, P.; Schweizer, J.; Ultsch, A.: A hybrid Expert System for Avalanche Forecasting, in publ.
Ultsch, A. „Konnektionistische Modelle und ihre Integration mit wissensbasierten Systemen“Habilitationsschrift Universität Dortmund, 1991.
Ultsch, A., Palm, G., Rückert, U.: Wissensverarbeitung in neuronaler Architektur, in: Brauer/Hernandez (Eds.): Verteilte künstliche Intelligenz und kooperatives Arbeiten, GI-Kongress, München, 1991, pp 508–518.
Ultsch, A. „Self-Organising Neural Networks for Visualisation and Classification“Proc. Conf. Soc. for Information and Classification, Dortmund Apr. 1992.
Ultsch, A. “Self-Organising Neural Networks for Monitoring and Knowledge Acquisition of a Chemical Process” Proc. ICANN-93, Amsterdam, p. 864–867.
Ultsch, A. & Halmans, G. “Neuronale Netze zur Unterstützung der Umweltforschung” Symp. Computer Science for Environmental Protection, Informatik Fachberichte 296, Springer, Dec. 1991.
Ultsch, A.& Li, H. “Automatic Acquisition of Symbolic Knowledge from Subsymbolic Neural Networks” Proc. IEEE Int. Conf. on Signal Processing, Peking, 1993.
Ultsch, A. & Siemon, H.P. „Exploratory Data Analysis: Using Kohonen Networks on Transputers“Fachberichte 329, FB Informatik, Universität Dortmund, Dec. 1989.
Wayand, M. “Anwendung von Kohonen Netzen zur Darstellung und Steuerung von chemischen Prozeßverläufen” Diplomarbeit, Univ. Dortmund 1992.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1994 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Ultsch, A., Guimaraes, G., Korus, D., Li, H. (1994). Knowledge Extraction from Artificial Neural Networks and Applications. In: Hektor, J., Grebe, R. (eds) Parallele Datenverarbeitung mit dem Transputer. Informatik aktuell. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-78901-4_11
Download citation
DOI: https://doi.org/10.1007/978-3-642-78901-4_11
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-57830-7
Online ISBN: 978-3-642-78901-4
eBook Packages: Springer Book Archive