Imputation of Missing Values by Inversion of Fuzzy Neuro-System

  • Krzysztof Siminski
Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 391)


Incomplete data are common and require special techniques. The essential techniques are: marginalisation, imputation, and rough sets. The paper presents the imputation by inversion of the neuro-fuzzy system. First the neuro-fuzzy systems is trained with complete data. Next the system is inverted and the missing values are imputed. The complete and imputed data are used to train the final neuro-fuzzy system. The technique is limited to data items with one missing value. The paper is accompanied by numerical examples and statistical verification.


Evolutionary optimisation Gradient descent Memetic algorithm Big-Bang-Big-crunch 


  1. 1.
    Acuña, E., Rodriguez, C.: The treatment of missing values and its effect on classifier accuracy. In: Banks, D., McMorris, F., Arabie, P., Gaul, W. (eds.) Classification, Clustering, and Data Mining Applications, pp. 639–647. Studies in Classification, Data Analysis, and Knowledge Organisation, Springer, Berlin (2004)Google Scholar
  2. 2.
    Batista, G.E.A.P.A., Monard, M.C.: An analysis of four missing data treatment methods for supervised learning. Appl. Artif. Intell. 17(5–6), 519–533 (2003)Google Scholar
  3. 3.
    Box, G.E.P., Jenkins, G.: Time Series Analysis Forecasting and Control. Holden-Day Incorporated, Oakland, California (1970)zbMATHGoogle Scholar
  4. 4.
    Chen, J.Q., Xi, Y.G., Zhang, Z.J.: A clustering algorithm for fuzzy model identification. Fuzzy Sets Syst. 98(3), 319–329 (1998)MathSciNetCrossRefGoogle Scholar
  5. 5.
    Cooke, M., Green, P., Josifovski, L., Vizinho, A.: Robust automatic speech recognition with missing and unreliable acoustic data. Speech Commun. 34, 267–285 (2001)CrossRefzbMATHGoogle Scholar
  6. 6.
    Czogala, E., Leski, J.: Fuzzy and Neuro-Fuzzy Intelligent Systems. Series in Fuzziness and Soft Computing, Physica-Verlag, A Springer-Verlag company, Heidelberg, New York (2000)Google Scholar
  7. 7.
    Filev, D.P.: Inversion of fuzzy models-practical issues. In: ICSFP, vol. 2, pp. 1658–1663. Anchorage, AK (1998)Google Scholar
  8. 8.
    Frank, A., Asuncion, A.: UCI machine learning repository (2010)Google Scholar
  9. 9.
    Gabriel, T.R., Berthold, M.R.: Missing values in fuzzy rule induction. In: SMC, vol. 2, pp. 1473–1476 (2005)Google Scholar
  10. 10.
    Galichet, S., Boukezzoula, R., Foulloy, L.: Explicit analytical formulation and exact inversion of decomposable fuzzy systems with singleton consequents. Fuzzy Sets Syst. 146, 421–436 (2004)MathSciNetCrossRefzbMATHGoogle Scholar
  11. 11.
    Grzymala-Busse, J., Goodwin, L., Grzymala-Busse, W., Zheng, X.: Handling missing attribute values in preterm birth data sets. In: Slezak, D., Yao, J., Peters, J., Ziarko, W., Hu, X. (eds.) Rough Sets, Fuzzy Sets, Data Mining, and Granular Computing. LNCS, vol. 3642, pp. 342–351. Springer, Berlin (2005)CrossRefGoogle Scholar
  12. 12.
    Grzymala-Busse, J.W.: On the unknown attribute values in learning from examples. In: Ras, Z., Zemankova, M. (eds.) Methodologies for Intelligent Systems. LNCS, vol. 542, pp. 368–377. Springer, Berlin (1991)CrossRefGoogle Scholar
  13. 13.
    Hathaway, R., Bezdek, J.: Fuzzy c-means clustering of incomplete data. IEEE Trans. Syst. Man Cybern. Part B: Cybern. 31(5), 735–744 (2001)CrossRefGoogle Scholar
  14. 14.
    Himmelspach, L., Conrad, S.: Fuzzy clustering of incomplete data based on cluster dispersion. In: Hüllermeier, E., Kruse, R., Hoffmann, F. (eds.) IPMU 2010. LNCS, vol. 6178, pp. 59–68. Springer, Berlin (2010)Google Scholar
  15. 15.
    Korytkowski, M., Nowicki, R., Scherer, R., Rutkowski, L.: Ensemble of rough-neuro-fuzzy systems for classification with missing features. In: FUZZ-IEEE, pp. 1745–1750. Hong Kong (2008)Google Scholar
  16. 16.
    Kumbasar, T., Eksin, İ., Güzelkaya, M., Yeşil, E.: Big bang big crunch optimization method based fuzzy model inversion. MICAI 2008: Advances in Artificial Intelligence. LNCS, pp. 737–740. Springer, Berlin (2008)Google Scholar
  17. 17.
    Kumbasar, T., Eksin, İ., Güzelkaya, M., Yeşil, E.: Exact inversion of decomposable interval type-2 fuzzy logic systems. Int. J. Approximate Reasoning 54, 253–272 (2013)CrossRefzbMATHGoogle Scholar
  18. 18.
    Matyja, A., Simiński, K.: Comparison of algorithms for clustering incomplete data. Found. Comput. Decis. Sci. 39(2), 107–127 (2014)Google Scholar
  19. 19.
    Mundfrom, D.J., Whitcomb, A.: Imputing missing values: the effect on the accuracy of classification. Multiple Linear Regres. Viewpoints 25(1), 13–19 (1998)Google Scholar
  20. 20.
    Nowicki, R.K.: Rough-neuro-fuzzy structures for classification with missing data. IEEE Trans. Syst. Man Cybern. Part B: Cybern. 39(6), 1334–1347 (2009)CrossRefGoogle Scholar
  21. 21.
    Ridders, C.: A new algorithm for computing a single root of a real continuous function. IEEE Trans. Circ. Syst. 26, 979–980 (1979)CrossRefzbMATHGoogle Scholar
  22. 22.
    Sikora, M., Simiński, K.: Comparison of incomplete data handling techniques for neuro-fuzzy systems. Comput. Sci. 15(4), 441–458 (2014)CrossRefGoogle Scholar
  23. 23.
    Sikora, M., Krzykawski, D.: Application of data exploration methods in analysis of carbon dioxide emission in hard-coal mines dewater pump stations. Mechanizacja i Automatyzacja Gornictwa 413(6) (2005)Google Scholar
  24. 24.
    Sikora, M., Sikora, B.: Application of machine learning for prediction a methane concentration in a coal-mine. Arch. Min. Sci. 51(4), 475–492 (2006)Google Scholar
  25. 25.
    Simiński, K.: Neuro-rough-fuzzy approach for regression modelling from missing data. Int. J. Appl. Math. Comput. Sci. 22(2), 461–476 (2012)zbMATHGoogle Scholar
  26. 26.
    Simiński, K.: Clustering with missing values. Fundamenta Informaticae 123(3), 331–350 (2013)zbMATHGoogle Scholar
  27. 27.
    Simiński, K.: Rough fuzzy subspace clustering for data with missing values. Comput. Inf. 33(1), 131–153 (2014)Google Scholar
  28. 28.
    Simiński, K.: Rough subspace neuro-fuzzy system. Fuzzy Sets Syst. 269, 30–46 (2015)CrossRefGoogle Scholar
  29. 29.
    Timm, H., Kruse, R.: Fuzzy cluster analysis with missing values. In: NAFIPS, pp. 242–246. Pensacola Beach, FL (1998)Google Scholar
  30. 30.
    Varkonyi-Koczy, A., Almos, A., Kovacshazy, T.: Genetic algorithms in fuzzy model inversion. In: FUZZ-IEEE, vol. 3, pp. 1421–1426 (1999)Google Scholar
  31. 31.
    Wagstaff, K.L., Laidler, V.G.: Making the most of missing values: object clustering with partial data in astronomy. In: ADASS XIV, vol. 347, pp. 172–176. Pasadena, California, USA (2005)Google Scholar
  32. 32.
    Xu, C., Shin, Y.: A fuzzy inverse model construction method for general monotonic multi-input-single-output (MISO) systems. IEEE Trans. Fuzzy Syst. 16(5), 1216–1231 (2008)CrossRefGoogle Scholar
  33. 33.
    Yeh, I.C.: Modeling of strength of high-performance concrete using artificial neural networks. Cement Concr. Res. 28(12), 1797–1808 (1998)CrossRefGoogle Scholar
  34. 34.
    Zhang, S.: Parimputation: from imputation and null-imputation to partially imputation. IEEE Intell. Inf. Bull. 9(1), 32–38 (2008)Google Scholar

Copyright information

© Springer International Publishing Switzerland 2016

Authors and Affiliations

  1. 1.Institute of InformaticsSilesian University of TechnologyGliwicePoland

Personalised recommendations