Abstract
We present our experience in visualization multivariate data when the data vectors have class assignment. The goal is then to visualize the data in such a way that data vectors belonging to different classes (subgroups) appear differentiated as much as possible. We consider for this purpose the traditional CDA (Canonical Discriminant Functions), the GDA 2.(Generalized Discriminant Analysis, Baudat and Anouar, 2000) and the Supervised SOM 7.(Kohonen, Makivasara, Saramaki 1984). The methods are applied to a set of 3-dimensional erosion data containing N=3420 data vectors subdivided into 5 classes of erosion risk. By performing the mapping of these data to a plane, we hope to gain some experience how the mentioned methods work in practice and what kind of visualization is obtained. The final conclusion is that the traditional CDA is the best both in speed (time) of the calculations and in the ability of generalization.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Bartkowiak, A., Szustalewicz, A., Evelpidou, N., Vassilopoulos, A.: Choosing data vectors representing a huge data set: a comparison of Kohonen’s maps and the neural gas method. Proc. of the First Int. Conf. on Environmental Research and Assessment, Bucharest, Romania, March 23–27, 2003. Ars Docendi Publishing House, Bucharest (2003) 5–20
Baudat, G., Anouar, F.: Generalized discriminant analysis using a kernel approach. Neural Computation 12 (2000) 2385–2404
Baudat G., Anouar F., Feature vector selection and projection using kernels. Neurocomputing 55 (2003) 21–38
Camastra, F., Vinciarelli, A.: Estimating the intrinsic dimension of data with a fractal-based method, IEEE Transactions on Pattern Analysis and Machine Intelligence, 24 (2002) 10 1404–1407
Duda, R.O., Hart, P.E., Stork D.E.: Pattern Classification, 2nd Edition, Wiley (2001)
Grassberger, P., Procaccia, I.: Measuring the strangeness of strange attractors. Physica D9 (1983) 189–208
Kohonen, T., Makisavara, K., Saramaki, T.: Phonetic maps — insightful representation of phonological features for speech recognition. ICPR Montreal, Canada (1984) 182–185
Shawe-Taylor, J., Christianini, N.: Kernel Methods for Pattern Recognition, Cambridge University Press (2004)
Scholkopf, B., Smola, A., Muller, K.R.: Nonlinear component analysis as a kernel eigen-value problem. Neural Computation 10 (1998) 1291–1398
Vesanto, J., Himberg, J., Alhoniemi, E., Parhankangas, J.: SOM Toolbox for Matlab 5. Som Toolbox Team, Helsinki, HUT, Finland, Libella Oy, Espoo (2000), http://www.cis.hut.fi/projects/somtoolbox (2005)
Webb, A., Statistical Pattern Recognition. 2nd Edition, Wiley (2002) reprint (2004)
Yang, J., Jin, Z., et. al.: Essence of kernel Fisher discriminant: KPCA plus LDA. Pattern Recognition 37 (2004) 2097–2100
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2006 Springer Science+Business Media, LLC
About this paper
Cite this paper
Bartkowiak, A., Evelpidou, N. (2006). Visualization of Some Multi-Class Erosion Data Using GDA and Supervised SOM. In: Saeed, K., Pejaś, J., Mosdorf, R. (eds) Biometrics, Computer Security Systems and Artificial Intelligence Applications. Springer, Boston, MA. https://doi.org/10.1007/978-0-387-36503-9_2
Download citation
DOI: https://doi.org/10.1007/978-0-387-36503-9_2
Publisher Name: Springer, Boston, MA
Print ISBN: 978-0-387-36232-8
Online ISBN: 978-0-387-36503-9
eBook Packages: Computer ScienceComputer Science (R0)