Pattern Discovery Through Separable Data Projections

  • Leon Bobrowski
  • Volodymir Mashtalir
  • Magdalena Topczewska
Part of the Advances in Soft Computing book series (AINSC, volume 45)


Data projections or, more generally, data linear transformations, in some cases allow to enhance interesting regularities in data sets. We pay particular attention to linear transformations from multidimensional feature space on a line and on a plane. In such cases, transformed data sets can be visualized and the resulting patterns can be evaluated by an expert both analytically and subjectively in accordance with the expert’s opinion. The projection pursuit provides well developed methods for designing interesting projections of data sets related to the normal model. Here we are considering separability criteria for designing projections.


Feature Vector Feature Space Criterion Function Exploratory Data Analysis Pattern Discovery 
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.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Johnson RA, Wichern DW (1991) Applied Multivariate Statistical Analysis, Prentice-Hall Inc., Englewood Cliffs, New York, 1991Google Scholar
  2. 2.
    Duda OR, Hart PE, Stork DG (1986) Pattern Classification, J. Wiley, NewGoogle Scholar
  3. 3.
    Jollife I (1986) Principal Component Analysis. Springer-Verlag, New. York, 1986Google Scholar
  4. 4.
    Lachenbruch P (1975) Discriminant analysis, New York: Hafner Press, first edition, 1975Google Scholar
  5. 5.
    Friedman JH, Tukey JW (1974) A projection pursuit algorithm for exploratory data analysis, IEEE Trans. of Computers, c-23(9):881–890, 1974.CrossRefGoogle Scholar
  6. 6.
    Bobrowski L (2005) Exploratory data analysis based on convex and piecewise linear (CPL) criterion functions (in Polish), Bialystok Technical University, 2005Google Scholar
  7. 7.
    Klein JP, Moeschberger ML (1997) Survival Analysis, Techniques for Censored and Truncated Data, Springer, NY 1997Google Scholar
  8. 8.
    Bobrowski L (2005) Ranked modeling with feature selection based on the CPL criterion functions, in Machine Learning and Data Mining in Pattern Recognition, Eds. P. Permer et al., Lecture Notes in Computer Science vol. 3587, Springer Verlag, Berlin 2005Google Scholar
  9. 9.
    Bobrowski L (1991) Design of piecewise linear classifiers from formal neurons by some basis exchange technique, Pattern Recognition, 24(9), pp. 863–870, 1991.CrossRefGoogle Scholar
  10. 10.
    Bobrowski L, Wasyluk H, (2001) Diagnosis supporting rules of the Hepar system, pp. 1309–1313 in: MEDINFO 2001, Proceedings of the 10-th World Congress on Medical Informatics, Ed. by V. L. Patel, R. Rogers, R Haux, IMIA, IOS Press Amsterdam 2001Google Scholar
  11. 11.
    Bobrowski L, Wasyluk H (2007) Induction of Diagnostic Support Rules through Data Mapping-on the Example of the Hepar system, Biocybernetics and Biomedical Engineering (in press)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2007

Authors and Affiliations

  • Leon Bobrowski
    • 1
  • Volodymir Mashtalir
    • 2
  • Magdalena Topczewska
    • 1
  1. 1.Faculty of Computer ScienceBialystok Technical UniversityPoland
  2. 2.Kharkov National University of Radio ElectronicsUkraine

Personalised recommendations