Abstract
A new algorithm named probabilistic polar self-organizing map (PPoSOM) is proposed. PPoSOM is a new variant of Polar SOM which is constructed on 2-D polar coordinates. Data weight and feature are represented by two variables that are radius and angle. The neurons on the map are set as data characteristic benchmarks. Projected data points are trained to get close to the neurons with the highest similarities, while weights of neurons are updated by a probabilistic data assignment method. Thus, not only similar data are gathered together, data characteristics are also reflected by their positions on the map. Our obtained results are compared with conventional SOM and ViSOM. The comparative results show that PPoSOM is a new effective method for multidimensional data visualization.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Johnson, R.A., Wichern, D.W.: Applied Multivariate Statistical Analysis. Prentice-Hall, Englewood Cliffs (1992)
Shepard, R.N., Carroll, J.D.: Parametric representation of nonlinear data structures. In: Krishnaiah, P.R. (ed.) Proc. Int. Symp. Multivariate Anal., pp. 561–592. Academic, New York (1965)
Kohonen, T.: Self-Organizing Maps. Springer, Berlin (1997)
Kohonen, T.: Self-organized formation of topologically correct feature maps. Biological Cybernetics 43, 59–69 (1982)
Vesanto, J., Alhoniemi, E.: Clustering of the self-organizing map. IEEE Trans. Neural Networks 11, 586–600 (2000)
Yin, H.: ViSOM: A novel method for multivariate data projection and structure visualization. IEEE Trans. Neural Networks 13, 237–243 (2002)
Yin, H.: Data visualization and manifold mapping using the ViSOM. IEEE Trans. Neural Networks 15, 1005–1016 (2002)
Yin, H.: Self-Organizing Maps: Statistical Analysis, Treatment and Applications, PhD Thesis, Department of Electronics, University of York, UK (1996)
Wu, S., Chow, T.W.S.: PRSOM: A New Visualization Method by Hybridizing Multidimensional Scaling and Self-Organizing Map. IEEE Trans. Neural Networks 16, 1362–1380 (2005)
Xu, L., Xu, Y., Chow, T.W.S.: PolSOM: A New Method for Multidimensional Data Visualization, submitted to Pattern Recognition
Fisher, R.A.: The use of multiple measure in taxonomic problems. Ann. Eugenics (Part II) 7, 179–188 (1936)
Newman, D.J., Hettich, S., Blake, C.L., Merz, C.J.: Uci Repository of Machine Learning Databases, University of California, Irvine, Dept. of Information and Computer Sciences (1998), http://archive.ics.uci.edu/ml/
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2009 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Xu, Y., Xu, L., Chow, T.W.S., Fong, A.S.S. (2009). PPoSOM: A Multidimensional Data Visualization Using Probabilistic Assignment Based on Polar SOM. In: Leung, C.S., Lee, M., Chan, J.H. (eds) Neural Information Processing. ICONIP 2009. Lecture Notes in Computer Science, vol 5863. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-10677-4_24
Download citation
DOI: https://doi.org/10.1007/978-3-642-10677-4_24
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-10676-7
Online ISBN: 978-3-642-10677-4
eBook Packages: Computer ScienceComputer Science (R0)