Performance evaluation of the self‐organizing map for feature extraction

[1] Despite its wide applications as a tool for feature extraction, the Self-Organizing Map (SOM) remains a black box to most meteorologists and oceanographers. This paper evaluates the feature extraction performance of the SOM by using artificial data representative of known patterns. The SOM is shown to extract the patterns of a linear progressive sine wave. Sensitivity studies are performed to ascertain the effects of the SOM tunable parameters. By adding random noise to the linear progressive wave data, it is demonstrated that the SOM extracts essential patterns from noisy data. Moreover, the SOM technique successfully chooses among multiple sets of patterns in contrast with an Empirical Orthogonal Function method that fails to do this. A practical way to apply the SOM is proposed and demonstrated using several examples, including long time series of coastal ocean currents from the West Florida Shelf. With improved SOM parameter choices, strong current patterns associated with severe weather forcing are extracted separate from previously identified asymmetric upwelling/downwelling and transitional patterns associated with more typical weather forcing.

[1]  Anthony J. Richardson,et al.  Using self-organizing maps to identify patterns in satellite imagery , 2003 .

[2]  Robert H. Weisberg,et al.  Patterns of ocean current variability on the West Florida Shelf using the self-organizing map , 2005 .

[3]  Robert G. Crane,et al.  Neural Nets: Applications in Geography , 1994 .

[4]  Francisco J. Ocampo-Torres,et al.  Seasonal and interannual variability of satellite-derived chlorophyll pigment, surface height, and temperature off Baja California , 2004 .

[5]  Anthony J. Richardson,et al.  Relating sardine recruitment in the Northern Benguela to satellite-derived sea surface height using a neural network pattern recognition approach , 2003 .

[6]  E. J. Ainsworth,et al.  Visualization of Ocean Colour and Temperature from multi-spectral imagery captured by the Japanese ADEOS satellite , 1999 .

[7]  Anthony J. Richardson,et al.  Identification and classification of vertical chlorophyll patterns in the Benguela upwelling system and Angola-Benguela front using an artificial neural network , 2001 .

[8]  B. Hewitson,et al.  Self-organizing maps: applications to synoptic climatology , 2002 .

[9]  Fouad Badran,et al.  Hierarchical clustering of self-organizing maps for cloud classification , 2000, Neurocomputing.

[10]  Kuolin Hsu,et al.  Self‐organizing linear output map (SOLO): An artificial neural network suitable for hydrologic modeling and analysis , 2002 .

[11]  Y. Hong,et al.  Self‐organizing nonlinear output (SONO): A neural network suitable for cloud patch–based rainfall estimation at small scales , 2005 .

[12]  William W. Hsieh,et al.  Nonlinear principal component analysis by neural networks , 2001 .

[13]  Dudley B. Chelton,et al.  Variability in satellite winds over the Benguela upwelling system during 1999–2000 , 2004 .

[14]  Esa Alhoniemi,et al.  SOM Toolbox for Matlab 5 , 2000 .

[15]  Y. Hong,et al.  Precipitation Estimation from Remotely Sensed Imagery Using an Artificial Neural Network Cloud Classification System , 2004 .

[16]  Gary S. E. Lagerloef,et al.  Empirical orthogonal function analysis of advanced very high resolution radiometer surface temperature patterns in Santa Barbara Channel , 1988 .

[17]  Esa Alhoniemi,et al.  Self-organizing map in Matlab: the SOM Toolbox , 1999 .

[18]  Melody Y. Kiang,et al.  An Evaluation of Self-Organizing Map Networks as a Robust Alternative to Factor Analysis in Data Mining Applications , 2001, Inf. Syst. Res..

[19]  Björn A. Malmgren,et al.  Climate Zonation in Puerto Rico Based on Principal Components Analysis and an Artificial Neural Network , 1999 .

[20]  F. Murtagh,et al.  The Kohonen self-organizing map method: An assessment , 1995 .

[21]  C. Basaran.  Preface , 1934, The Yale Journal of Biology and Medicine.

[22]  Frank E. Muller-Karger,et al.  West Florida Shelf Response to Local Wind Forcing: April 1998 , 2001 .

[23]  T. Kohonen,et al.  Bibliography of Self-Organizing Map SOM) Papers: 1998-2001 Addendum , 2003 .

[24]  Teuvo Kohonen,et al.  The self-organizing map , 1990, Neurocomputing.

[25]  Ruoying He,et al.  Sea Surface Temperature Patterns on the West Florida Shelf Using Growing Hierarchical Self-Organizing Maps , 2006 .

[26]  Samuel Kaski,et al.  Bibliography of Self-Organizing Map (SOM) Papers: 1981-1997 , 1998 .

[27]  E. J. Ainsworth,et al.  Radiance spectra classification from the Ocean Color and Temperature Scanner on ADEOS , 1999, IEEE Trans. Geosci. Remote. Sens..

[28]  William W. Hsieh,et al.  Nonlinear multivariate and time series analysis by neural network methods , 2004 .

[29]  B. Weare,et al.  Empirical Orthogonal Analysis of Pacific Sea Surface Temperatures , 1976 .

[30]  M. V. Velzen,et al.  Self-organizing maps , 2007 .

[31]  John G. Field,et al.  Identifying characteristic chlorophyll a profiles in the coastal domain using an artificial neural network , 2002 .

[32]  Teuvo Kohonen,et al.  Self-organized formation of topologically correct feature maps , 2004, Biological Cybernetics.

[33]  Diana Liverman,et al.  Intraseasonal Variability Associated with Wet Monsoons in Southeast Arizona , 2002 .

[34]  John M. Klinck,et al.  EOF Analysis of Central Drake Passage Currents from DRAKE 79 , 1985 .

[35]  Tereza Cavazos.  Using Self-Organizing Maps to Investigate Extreme Climate Events: An Application to Wintertime Precipitation in the Balkans , 2000 .

[36]  T. Kohonen.  Self-Organized Formation of Correct Feature Maps , 1982 .