Probabilistic Principal Surface Classifier

  • Kuiyu Chang
  • Joydeep Ghosh
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3614)

Abstract

In this paper we propose using manifolds modeled by probabilistic principle surfaces (PPS) to characterize and classify high-D data. The PPS can be thought of as a nonlinear probabilistic generalization of principal components, as it is designed to pass through the “middle” of the data. In fact, the PPS can map a manifold of any simple topology (as long as it can be described by a set of ordered vector co-ordinates) to data in high-dimensional space. In classification problems, each class of data is represented by a PPS manifold of varying complexity. Experiments using various PPS topologies from a 1-D line to 3-D spherical shell were conducted on two toy classification datasets and three UCI Machine Learning datasets. Classification results comparing the PPS to Gaussian Mixture Models and K-nearest neighbours show the PPS classifier to be promising, especially for high-D data.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Hastie, T., Stuetzle, W.: Principal curves. Journal of the American Statistical Association 84, 502–516 (1988)MathSciNetCrossRefGoogle Scholar
  2. 2.
    Mulier, F., Cherkassky, V.: Self-organization as an iterative kernel smoothing process. Neural Computation 7, 1165–1177 (1995)CrossRefGoogle Scholar
  3. 3.
    Kohonen, T.: Self-Organizing Maps. Springer, Heidelberg (1995)Google Scholar
  4. 4.
    Jain, A.K., Mao, J.: Artificial neural network for nonlinear projection of multivariate data. In: IEEE IJCNN, Baltimore, MD, vol. 3, pp. 335–340 (1992)Google Scholar
  5. 5.
    Mao, J., Jain, A.K.: Artificial neural networks for feature extraction and multivariate data projection. IEEE Transactions on Neural Networks 6(2), 296–317 (1995)CrossRefGoogle Scholar
  6. 6.
    Friedman, J.H.: An overview of predictive learning and function approximation. In: Cherkassky, V., Friedman, J.H., Wechsler, H. (eds.) Proc. NATO/ASI Workshop From Statistics to Neural Networks, pp. 1–61. Springer Verlag, Heidelberg (1994)Google Scholar
  7. 7.
    Bishop, C.M.: Neural Networks for Pattern Recognition, 1st edn. Clarendon Press, Oxford (1995)Google Scholar
  8. 8.
    Chang, K.Y., Ghosh, J.: A unified model for probabilistic principal surfaces. IEEE Transactions on Pattern Analysis and Machine Intelligence 23, 22–41 (2001)CrossRefGoogle Scholar
  9. 9.
    Bishop, C.M., Svensén, M., Williams, C.K.I.: GTM: The generative topographic mapping. Neural Computation 10, 215–235 (1998)CrossRefGoogle Scholar
  10. 10.
    Bishop, C.M., Svensén, M., Williams, C.K.I.: Developments of the generative topographic mapping. Neurocomputing 21, 203–224 (1998)MATHCrossRefGoogle Scholar
  11. 11.
    Chang, K.Y., Ghosh, J.: Principal curve classifier – a nonlinear approach to pattern classification. In: International Joint Conference on Neural Networks, Anchorage, Alaska, USA, pp. 695–700. IEEE, Los Alamitos (1998)Google Scholar
  12. 12.
    Aviles-Cruz, C., Guérin-Dugué, A., Voz, J.L., Cappel, D.V.: Enhanced learning for evolutive neural architecture. Technical Report Deliverable R3-B1-P, INPG, UCL, TSA (1995)Google Scholar
  13. 13.
    Blake, C., Merz, C.: UCI repository of machine learning databases (1998)Google Scholar
  14. 14.
    Chang, K.Y., Ghosh, J.: Three-dimensional model-based object recognition and pose estimation using probabilistic principal surfaces. In: SPIE:Applications of Artificial Neural Networks in Image Processing V, San Jose, California, USA. SPIE, vol. 3962, pp. 192–203 (2000)Google Scholar
  15. 15.
    Staiano, A., Tagliaferri, R., Vinco, L.D.: High-d data visualization methods via probabilistic principal surfaces for data mining applications. In: International Workshop on Multimedia Databases and Image Communication, Salerno, Italy (2004)Google Scholar
  16. 16.
    Staiano, A., Vinco, L.D., Ciaramella, A., Raiconi, G., Tagliaferri, R., Longo, G., Miele, G., Amato, R., Mondo, C.D., Donalek, C., Mangano, G., Bernardo, D.D.: Probabilistic principal surfaces for yeast gene microarray data mining. In: International Conference on Data Mining, pp. 202–208 (2004)Google Scholar
  17. 17.
    Staiano, A., Tagliaferri, R., Longo, G., Benvenuti, P.: Committee of spherical probabilistic surfaces. In: International Joint Conference on Neural Networks, Budapest, Hungary (2004)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2005

Authors and Affiliations

  • Kuiyu Chang
    • 1
  • Joydeep Ghosh
    • 2
  1. 1.School of Computer EngineeringNanyang Technological UniversitySingaporeSingapore
  2. 2.Department of Electrical and Computer EngineeringUniversity of Texas at AustinAustinUSA

Personalised recommendations