Advertisement

Fuzzy Control pp 109-120 | Cite as

Adaptive Information System for Data Approximation Problems

  • Witold Kosiński
  • Martyna Weigl
Conference paper
Part of the Advances in Soft Computing book series (AINSC, volume 6)

Abstract

An adaptive information system is constructed for the approximation of a multidimensional data base. Such a construction problem can be regarded as an approximation of a multivariate real-valued function that is known in a discrete number of points. That set of points forming the multidimensional data base is called a training set TRE for the adaptive information system. The constructed system contains a module of one-conditional fuzzy rules consequent parts of which are artificial feed-forward neural networks. The numerical data contained in TRE are used to construct premise parts of each rule and membership functions of fuzzy sets involved in them. They are specified with the help of a clustering analysis of TRE. For each fuzzy rule the neural network appearing as its consequent part is trained on the corresponding cluster as on its subdomain In that way the cluster analysis can supply a knowledge in designing the approximator.

Keywords

Membership Function Fuzzy Rule Inference System Fuzzy Inference System Consequent Part 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Anderberg, M.R.: Cluster Analysis for Applications. Probability and Mathematical Statistics, Academic Press, New York (1973)Google Scholar
  2. 2.
    Buckley, J. J. and Hayashi, Y.: Numerical relationships between neural networks, continuous functions, and fuzzy systems. Fuzzy Sets and Systems, 60 (1993) 1–8MathSciNetCrossRefGoogle Scholar
  3. 3.
    Cybenko, G.: Approximation by superpositions of sigmoidal function. Mathematics of Control, Signals, and Systems, 2 (1989) 303–314MathSciNetMATHCrossRefGoogle Scholar
  4. 4.
    Funahashi, K.: On the approximate realization of continuous mapping by neural networks. Neural Networks, 2 (1989) 183–192CrossRefGoogle Scholar
  5. 5.
    Hand, D.J.: Discrimination and Classification. Wiley, Chichester (1981)MATHGoogle Scholar
  6. 6.
    Golg,bek, B., Kosinski, W. and Weigl, M.: Adaptation of learning rate via adaptation of weight vector in modified M-Delta networks. In Computational Intelligence and Applications, P.S. Szczepaniak (Ed.), (Studies in Fuzziness and Soft Computing, Vol. 23), Physics-Verlag, c/o Springer-Verlag, (1999) 156–163Google Scholar
  7. 7.
    Hornik, K.: Approximation capabilities of multilayer feedforward networks. Neural Networks, 4 (1991) 251–257CrossRefGoogle Scholar
  8. 8.
    Jang, J. and Sun, T-C.: Functional equivalence between radial basis function networks and fuzzy inference system. IEEE Transaction on Neural Networks, 4 (1), (1993) 156–159CrossRefGoogle Scholar
  9. 9.
    Kies, P., Kosinski, W. and Weigl, M.: Seed growing approach in clustering analysis. In Intelligent Information Systems. Proceedings of the VI-th International Workshop on Intelligent Information Systems, Zakopane, 9–13 June, 1997, Instytut Podstaw Informatyki, PAN, Warszawa (1997) 7–15Google Scholar
  10. 10.
    Kolesnik, R., Kole§nik, L. and Kosinski, W.: Genetic operators for clustering analysis, In Intelligent Information Systems VIII, Proc. of the Workshop held in Ustron, Poland, 14–18 June, 1999, Instytut Podstaw Informatyki, PAN, Warszawa (1999) 203–208Google Scholar
  11. 11.
    Kosinski, W., Weigl, M. and Kies, P.: Fuzzy domain covering of an inference system. In Proceedings of the Third National Conference on Neural Networks and Their Applications, October 14–18, 1997, Kule n. Czestochowa, PNNS, Czestochowa (1997) 310–321Google Scholar
  12. 12.
    Kosinski, W. and Weigl, M.: General mapping approximation problems solving by neural networks and fuzzy inference systems. Systems Analysis Modelling Simulation, 30 (1), (1998) 11–28MATHGoogle Scholar
  13. 13.
    Kosinski, W., Weigl, M., and Michalewicz Z.: Evolutionary domain covering of an inference system for function approximation. In Evolutionary Programming VII V. W. Port, N. Saravanam, D. Waagen and A. E. Eiben (Eds.), Proceedings of the 7th International Conference, EP’98, San Diego, California, USA, March 25–27, 1998, Lecture Notes in Computer Science, Vol. 1447, New York (1998) 167–180CrossRefGoogle Scholar
  14. 14.
    Kosinski W. Weigl M., Michalewicz Z. and Kolesnik R. Genetic algorithms for preprocessing of data for universal approximators. In Intelligent Information Systems VII, Proceedings of the Workshop held in Malbork, Poland, 15–19 June, 1998, Instytut Podstaw Informatyki, PAN, Warszawa (1998) 320–331Google Scholar
  15. 15.
    Kowalczyk, D.: An adaptive fuzzy inference system as a fuzzy controller (Rozmyty adaptacyjny system wnioskujg,cy jako sterownik rozmyty). Master Thesis, (in Polish), WSP w Bydgoszczy, Wydzial Matematyki, Techniki i Nauk Przyrodniczych, Specjalnosé: Technika Komputerowa, Bydgoszcz, czerwiec (1999)Google Scholar
  16. 16.
    Mhaskar, H. and Micchelli, C.: Approximation by superposition of a sigmoidal function and radial basis functions. Advances Applied Mathematics, 13 (1992) 350–373MathSciNetMATHCrossRefGoogle Scholar
  17. 17.
    Mhaskar, H. and Micchelli, C.: Dimension-independent bounds on the degree of approximation by neural network. IBM Journal of Research and Development, 38 (3) (1994) 277–284MATHCrossRefGoogle Scholar
  18. 18.
    Scarselli, F. and Tsoi, A. Ch.: Universal approximation using feedforward neural networks: A survey of some existing methods, and some new results. Neural Networks, 11 (1) (1998) 15–37CrossRefGoogle Scholar
  19. 19.
    Takagi, H. and Sugeno, M.: Fuzzy identification of systems and its applications to modeling and control. IEEE Trans. on Systems, Man. and Cybernetics, 15 (1985) 116–132MATHCrossRefGoogle Scholar
  20. 20.
    Weigl, M. and Kosinski, W.: Fuzzy inference system and modified back-propagation network in approximation problems. Proceedings of the III-rd International Symposium on Intelligent Information Systems, Wigry n. Suwalki, June 1994, Instytut Podstaw Informatyki, PAN, Warszawa (1994) 427–442Google Scholar
  21. 21.
    Weigl, M. Neural networks and fuzzy inference systems in approximation problems (Sieci neuronowe i rozmyte systemy wnioslcujgce w problemach aproksymacji). Ph. D. Thesis, (in Polish), IPPT PAN, Warszawa, czerwiec (1995)Google Scholar
  22. 22.
    Weigl, M. and Kosinski, W.: Fuzzy reasoning in adaptive expert systems for approximation problems. In Proceedings of the 3-th Zittau Fuzzy-Colloquy Zittau, September 5–6, 1995, Wissenschaftliche Berichte, Heft 41, (1995) 163–174Google Scholar
  23. 23.
    Weigl, M. and Kosinski, W.: Approximation of multivariate functions by generalized adaptive fuzzy inference network. In Proceedings of the 9-th International Symposium on Methodologies for Intelligent Systems ISMIS’96, Zakopane, June, 1996, IPI PAN, Warszawa (1996) 120–133Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2000

Authors and Affiliations

  • Witold Kosiński
    • 1
    • 2
    • 3
  • Martyna Weigl
    • 2
    • 4
  1. 1.Polish-Japanese Institute of Information TechnologyWarszawaPoland
  2. 2.Center of Mechanics and Information Technology Inst. Fund. Technol. Research, Polish Academy of SciencesIPPT PANWarszawaPoland
  3. 3.Department of Mathematics, Technology and Natural SciencesPedagogical University of BydgoszczBydgoszczPoland
  4. 4.Polska Telefonia CyfrowaERA GSMWarszawaPoland

Personalised recommendations