Comparison of Neural Network Boolean Factor Analysis Method with Some Other Dimension Reduction Methods on Bars Problem

  • Dušan Húsek
  • Pavel Moravec
  • Václav Snášel
  • Alexander Frolov
  • Hana Řezanková
  • Pavel Polyakov
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4815)


In this paper, we compare performance of novel neural network based algorithm for Boolean factor analysis with several dimension reduction techniques as a tool for feature extraction. Compared are namely singular value decomposition, semi-discrete decomposition and non-negative matrix factorization algorithms, including some cluster analysis methods as well. Even if the mainly mentioned methods are linear, it is interesting to compare them with neural network based Boolean factor analysis, because they are well elaborated. Second reason for this is to show basic differences between Boolean and linear case. So called bars problem is used as the benchmark. Set of artificial signals generated as a Boolean sum of given number of bars is analyzed by these methods. Resulting images show that Boolean factor analysis is upmost suitable method for this kind of data.


Singular Value Decomposition Singular Vector Dimension Reduction Technique Singular Value Decomposition Method Boolean Matrix Multiplication 
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.


  1. 1.
    Frolov, A.A., Sirota, A.M., Húsek, D., Muravjev, P.: Binary factorization in Hopfield-like neural networks: single-step approximation and computer simulations. In: Neural Networks World, pp. 139–152 (2004)Google Scholar
  2. 2.
    Frolov, A.A., Húsek, D., Muravjev, P., Polyakov, P.: Boolean Factor Analysis by Attractor Neural Network. Neural Networks, IEEE Transactions 18(3), 698–707 (2007)CrossRefGoogle Scholar
  3. 3.
    Földiák, P.: Forming sparse representations by local anti-Hebbian learning. Biological cybernetics 64(22), 165–170 (1990)CrossRefGoogle Scholar
  4. 4.
    Fierro, R.D., Jiang, E.P.: Lanczos and the Riemannian SVD in information retrieval applications. Numerical Algebra with Applications 3(1), 1–18 (2003)Google Scholar
  5. 5.
    Berry, M., Dumais, S., Letsche, T.: Computational Methods for Intelligent Information Access. In: Proceedings of the 1995 ACM/IEEE Supercomputing Conference, San Diego, California, USA (1995)Google Scholar
  6. 6.
    Berry, M.W., Fierro, R.D.: Low-Rank Orhogonal Decomposition for Information Retrieval Applications. Numerical Algebra with Applications 1(1), 1–27 (1996)Google Scholar
  7. 7.
    Shahnaz, F., Berry, M., Pauca, P., Plemmons, R.: Document clustering using nonnegative matrix factorization. Journal on Information Processing and Management 42, 373–386 (2006)zbMATHCrossRefGoogle Scholar
  8. 8.
    Kolda, T.G., O’Leary, D.P.: Computation and uses of the semidiscrete matrix decomposition. ACM Transactions on Information Processing  (2000)Google Scholar
  9. 9.
    Turk, M., Pentland, A.: Eigenfaces for recognition. Journal of Cognitive Neuroscience 3(1), 71–86 (1991)CrossRefGoogle Scholar
  10. 10.
    Spratling, M.W.: Learning Image Components for Object Recognition. Journal of Machine Learning Research 7, 793–815 (2006)MathSciNetGoogle Scholar
  11. 11.
    Frolov, A.A., Husek, D., Polyakov, P., Rezankova, H.: New Neural Network Based Approach Helps to Discover Hidden Russian Parliament Votting Paterns. In: Proceedings of International Joint Conference on Neural Networks, pp. 6518–6523. Omnipress, Madison, WI, USA (2006)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2007

Authors and Affiliations

  • Dušan Húsek
    • 1
  • Pavel Moravec
    • 2
  • Václav Snášel
    • 2
  • Alexander Frolov
    • 3
  • Hana Řezanková
    • 4
  • Pavel Polyakov
    • 5
  1. 1.Institute of Computer Science, Dept. of Neural Networks, Academy of Sciences of Czech Republic, Pod Vodárenskou věží 2, 182 07 PragueCzech Republic
  2. 2.Department of Computer Science, FEECS, VŠB – Technical University of Ostrava, 17. listopadu 15, 708 33 Ostrava-PorubaCzech Republic
  3. 3.Institute of Higher Nervous Activity and Neurophysiology, Russian Academy of Sciences, Butlerova 5a, 117 485 MoscowRussia
  4. 4.Department of Statistics and Probability, University of Economics, Prague, W. Churchill sq. 4, 130 67 PragueCzech Republic
  5. 5.Institute of Optical Neural Technologies, Russian Academy of Sciences, Vavilova 44, 119 333 MoscowRussia

Personalised recommendations