Tips for Processing and Color-coding of Self-Organizing Maps

  • Samuel Kaski
  • Teuvo Kohonen
Part of the Springer Finance book series (FINANCE)


The most important aspects for creating good SOM maps are the selection of the size and shape of the SOM, the scaling of the input variables, the selection of the neighborhood function and the learning rate, and the initialization of the model vectors. In addition, we discuss automatic procedures for color-coding of SOM maps. The data entries corresponding to, for example, different countries can be automatically colored so that similar data attain a similar color.


Color Center Model Vector World Development Indicator Neighborhood Function Cluster Tendency 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer-Verlag Berlin Heidelberg 1998

Authors and Affiliations

  • Samuel Kaski
  • Teuvo Kohonen

There are no affiliations available

Personalised recommendations