Advertisement

New Measure of Boolean Factor Analysis Quality

  • Alexander A. Frolov
  • Dusan Husek
  • Pavel Yu. Polyakov
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6593)

Abstract

Learning of objects from complex patterns is a long-term challenge in philosophy, neuroscience, machine learning, data mining, and in statistics. There are some approaches in literature trying to solve this difficult task consisting in discovering hidden structure of high-dimensional binary data and one of them is Boolean factor analysis. However there is no expert independent measure for evaluating this method in terms of the quality of solutions obtained, when analyzing unknown data. Here we propose information gain, model-based measure of the rate of success of individual methods. This measure presupposes that observed signals arise as Boolean superposition of base signals with noise. For the case whereby a method does not provide parameters necessary for information gain calculation we introduce the procedure for their estimation. Using an extended version of the ”Bars Problem” generation of typical synthetics data for such a task, we show that our measure is sensitive to all types of data model parameters and attains its maximum, when best fit is achieved.

Keywords

Boolean factor analysis information gain Hopfield neural network statistics expectation-maximization associative memory neural network application Boolean matrix factorization bars problem 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Barlow, H.B.: Cerebral cortex as model builder. In: Rose, D., Dodson, V.G. (eds.) Models of the Visual Cortex, pp. 37–46. Wiley, Chichester (1985)Google Scholar
  2. 2.
    Belohlavek, R., Vychodil, V.: On Boolean factor analysis with formal concepts as factors, pp. 20–24 (2006)Google Scholar
  3. 3.
    Belohlavek, R., Vychodil, V.: Discovery of optimal factors in binary data via a novel method of matrix decomposition. Journal of Computer and System Sciences 76(1), 3–20 (2010)MathSciNetCrossRefzbMATHGoogle Scholar
  4. 4.
    Foldiak, P.: Forming sparse representations by local anti-hebbian learning. Biological Cybernetics 64, 165–170 (1990)CrossRefGoogle Scholar
  5. 5.
    Frolov, A.A., Husek, D., Muraviev, I.P., Polyakov, P.Y.: Boolean factor analysis by attractor neural network. IEEE Transactions on Neural Networks 18(3), 698–707 (2007)CrossRefGoogle Scholar
  6. 6.
    Frolov, A.A., Husek, D., Polyakov, P., Rezankova, H.: New Neural Network Based Approach Helps to Discover Hidden Russian Parliament Voting Patterns. In: IEEE International Joint Conference on Neural Networks, pp. 6518–6523 (2006)Google Scholar
  7. 7.
    Frolov, A.A., Husek, D., Polyakov, P.Y.: Recurrent neural network based Boolean factor analysis and its application to automatic terms and documents categorization. IEEE Transactions on Neural Networks 20(7), 1073–1086 (2009)CrossRefGoogle Scholar
  8. 8.
    Frolov, A.A., Husek, D., Polyakov, P.Y.: Origin and Elimination of Two Global Spurious Attractors in Hopfield-like Neural Network Performing Boolean Factor Analysis. Neurocomputing (2010) (in press)Google Scholar
  9. 9.
    Frolov, A.A., Húsek, D., Rezanková, H., Snásel, V., Polyakov, P.: Clustering variables by classical approaches and neural network Boolean factor analysis. In: IEEE International Joint Conference on Neural Networks, pp. 3742–3746 (2008)Google Scholar
  10. 10.
    Lücke, J., Sahani, M.: Maximal causes for non-linear component extraction. The Journal of Machine Learning Research 9, 1227–1267 (2008)MathSciNetzbMATHGoogle Scholar
  11. 11.
    Marr, D.: A Theory for Cerebral Neocortex. Proceedings of the Royal Society of London. Series B, Biological Sciences (1934-1990) 176(1043), 161–234 (1970)CrossRefGoogle Scholar
  12. 12.
    Marr, D.: Simple Memory: A Theory for Archicortex. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences (1934-1990) 262(841), 23–81 (1971)CrossRefGoogle Scholar
  13. 13.
    Mickey, M.R., Mundle, P., Engelman, L.: Boolean factor analysis. In: Dixon, W. (ed.) BMDP Statistical Software, pp. 538–545. University of California Press, Berkeley (1983)Google Scholar
  14. 14.
    Pelaez, J.R.: Plato’s theory of ideas revisited. Neural Networks 10(7), 1269–1288 (1997)CrossRefzbMATHGoogle Scholar
  15. 15.
    Veiel, H.O.: Psychopathology and Boolean Factor Analysis: a mismatch. Psychol. Med. 15(3), 623–628 (1985)CrossRefGoogle Scholar
  16. 16.
    Weber, A.C., Scharfetter, C.: The syndrome concept: history and statistical operationalizations. Psychol. Med. 14(2), 315–325 (1984)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • Alexander A. Frolov
    • 1
  • Dusan Husek
    • 2
  • Pavel Yu. Polyakov
    • 3
    • 4
  1. 1.Institute of Higher Nervous Activity and NeurophysiologyRussian Academy of SciencesMoscowRussia
  2. 2.Institute of Computer ScienceAcademy of Sciences of the Czech RepublicPrahaCzech Republic
  3. 3.Scientific-Research Institute for System StudiesRussian Academy of SciencesMoscowRussia
  4. 4.VŠB – Technical University of OstravaPorubaCzech Republic

Personalised recommendations