A PATTERN SELECTION ALGORITHM IN KERNEL PCA APPLICATIONS

Ruixin Yang, John Tan, Menas Kafatos

Abstract

Principal Component Analysis (PCA) has been extensively used in different fields including earth science for spatial pattern identification. However, the intrinsic linear feature associated with standard PCA prevents scientists from detecting nonlinear structures. Kernel-based principal component analysis (KPCA), a recently emerging technique, provides a new approach for exploring and identifying nonlinear patterns in scientific data. In this paper, we recast KPCA in the commonly used PCA notation for earth science communities and demonstrate how to apply the KPCA technique into the analysis of earth science data sets. In such applications, a large number of principal components should be retained for studying the spatial patterns, while the variance cannot be quantitatively transferred from the feature space back into the input space. Therefore, we propose a KPCA pattern selection algorithm based on correlations with a given geophysical phenomenon. We demonstrate the algorithm with two widely used data sets in geophysical communities, namely the Normalized Difference Vegetation Index (NDVI) and the Southern Oscillation Index (SOI). The results indicate the new KPCA algorithm can reveal more significant details in spatial patterns than standard PCA.

References

  1. Cox, T. F. and Cox, M. A. (2000). Multidimensional Scaling. Chapman & Hall.
  2. Lorenz, E. N. (1959). Empirical orthogonal functions and statistical weather prediction. Final Report, Statistical Forecasting Project, 1959; Massachusetts Institute of Technology, Dept. of Meteorology, 29-78.
  3. Mika, S., Schölkopf, B., Smola, A., M üller, K.-R., Scholz, M., and Rätsch, G. (1999). Kernel PCA and Denoising in Feature Spaces. In Kearns, M. S., Solla, S. A., and Cohn, D. A., editors, Advances in Neural Information Processing Systems 11, pages 536 - 542, Cambridge, MA. MIT Press.
  4. Monahan, A. H. (2001). Nonlinear Principal Component Analysis: Tropical IndoPacific Sea Surface Temperature and Sea Level Pressure. Journal of Climate, 14(2):219-233.
  5. NDMC (National Drought Mitigation Center) (2006). What is Drought? http://www.drought.unl.edu/whatis/elnino.htm (Last accessed on Feb. 8, 2006).
  6. Philander, S. G. (1990). El Nin˜o, La Nin˜a, and the Southern Oscillation. Academic Press.
  7. Ropelewski, C. and Jones, P. (1987). An Extension of the Tahiti-Darwin Southern Oscillation Index. Monthly Weather Review, 115(9):2161-2165.
  8. Roweis, S. and Saul, L. (2000). Nonlinear Dimensionality Reduction by Locally Linear Embedding. Science, 290(22 December 2000):2323-2326.
  9. Schölkopf, B., Burges, C., and Smola, J. (1999a). Advances in Kernel Methods: Support Vector Learning. MIT Press.
  10. Schölkopf, B., Mika, S., Burges, C., Knirsch, P., M üller, K.- R., Rätsch, G., and Smola, A. (1999b). Input Space vs. Feature Space in Kernel-Based Methods. IEEE Transactions on Neural Networks, 10(5):1000-1017.
  11. Schölkopf, B., Smola, A., and M üller, K.-R. (1998). Nonlinear Component Analysis as a Kernel Eigenvalue Problem. Neural Computation, 10(5):1299-1319.
  12. Tan, J. (2005). Applications of Kernel PCA Methods to Geophysical Data. George Mason University. PhD Thesis.
  13. Tan, J., Yang, R., and Kafatos, M. (2006). Kernel PCA Analysis for Remote Sensing Data. In 18th Conference on Climate Variability and Change. American Meteorological Society. Paper P1.5, Altanta, GA, CDROM.
  14. Tenenbaum, J., de Silva, V., and Langford, J. (2000). A Global Geometric Framework for Nonlinear Dimensionality Reduction. Science, 290(22 December 2000):2319-2323.
  15. Thompson, D. W. J. and Wallace, J. M. (2000). Annular Modes in the Extratropical Circulation. Part I: Monthto-Month Variability. Journal of Climate, 13(5):1000- 1016.
  16. Von Storch, H. and Zwiers, F. W. (1999). Statistical Analysis in Climate Research. Cambridge University Press.
  17. Wallace, J. M., Smith, C., and Bretherton, C. S. (1992). Singular Value Decomposition of Wintertime Sea Surface Temperature and 500-mb Height Anomalies. Journal of Climate, 5(6):561-576.
Download


Paper Citation


in Harvard Style

Yang R., Tan J. and Kafatos M. (2006). A PATTERN SELECTION ALGORITHM IN KERNEL PCA APPLICATIONS . In Proceedings of the First International Conference on Software and Data Technologies - Volume 2: ICSOFT, ISBN 978-972-8865-69-6, pages 195-202. DOI: 10.5220/0001320801950202


in Bibtex Style

@conference{icsoft06,
author={Ruixin Yang and John Tan and Menas Kafatos},
title={A PATTERN SELECTION ALGORITHM IN KERNEL PCA APPLICATIONS},
booktitle={Proceedings of the First International Conference on Software and Data Technologies - Volume 2: ICSOFT,},
year={2006},
pages={195-202},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0001320801950202},
isbn={978-972-8865-69-6},
}


in EndNote Style

TY - CONF
JO - Proceedings of the First International Conference on Software and Data Technologies - Volume 2: ICSOFT,
TI - A PATTERN SELECTION ALGORITHM IN KERNEL PCA APPLICATIONS
SN - 978-972-8865-69-6
AU - Yang R.
AU - Tan J.
AU - Kafatos M.
PY - 2006
SP - 195
EP - 202
DO - 10.5220/0001320801950202