Towards High-resolution Self-organizing Maps of Geographic Features - Geographic Visualization: Concepts, Tools and Applications - page 180

principally with SOMs of high and extremely high resolution that are as yet undefined.
Systematic studies should be undertaken to explore this, which may in particular call for
well-controlled synthetic data sets.
The use of climate data in a high-resolution SOM and the use of existing geographic objects
as place holders of climate attributes was informed by the desire to extend the notion of
attribute space travel (Skupin, 2007). One of the overarching goals of that research direction
is to provide a methodological framework for dealing with the experience of geographic
space in a computational manner. Socio-economic characteristics of a geographic place
have an effect on one's experience with and interaction in that place, but the physical
attributes, such as temperature and humidity, obviously play an important role as well. The
experimental work presented in this chapter is meant as a first step towards an integrated
visual modelling of physical and social attributes, in this case by attaching climate attributes
to demographic enumeration units. Future work will include the actual combination of very
different types of attributes, including climate, demographic, land use/land cover, and many
others, and thus create rich spatializations of geographic objects. Such work may include
irregular enumeration units - therefore incurring the cartogram effects described in this
chapter - as well as regular tessellations of geographic space.
Acknowledgements
We gratefully acknowledge the assistance of Charles Schmidt in the pre-processing of
climate data and of Martin Lacayo in creating geometric and attribute base data for the
climate SOM.
References
Deboeck, G. and Kohonen, T. (eds) (1998) Visual Explorations in Finance with Self-Organizing
Maps. Berlin, Springer.
Fayyad, U. M., Piatetsky-Shapiro, G., Smythe, P. and Uthurusamy, R. (eds) (1996) Advances in
Knowledge Discovery and Data Mining. Menlo Park, CA, AAAI/MIT Press.
Fritzke, B. (1999) Growing self-organizing networks - history, status quo, and perspectives. In
Kohonen Maps , Oja, E. and Kaski, S. (eds). Amsterdam, Elsevier.
Kohonen, T. (1982) Self-organized formation of topologically correct feature maps. Biological
Cybernetics 43: 59-69.
Kohonen, T. (1990) The self-organizing map. Proceedings of the IEEE 78: 1464-1480.
Kohonen, T. (2001) Self-Organizing Maps. Berlin, Springer.
Kohonen, T., Hynninen, J., Kangas, J. and Laaksonen, J. (1996a) SOM PAK: The Self-Organizing
Map Program Package. Espoo, Helsinki University of Technology, Laboratory of Computer
and Information Science.
Kohonen, T., Kaski, S., Lagus, K. and Honkela, T. (1996b) Very large two-level SOM for the
browsing of newsgroups. 1996 International Conference on Artificial Neural Networks. Springer,
Berlin.
Oja, E. and Kaski, S. (eds) (1999) Kohonen Maps. Amsterdam, Elsevier.
Sangole, A. and Knopf, G. K. (2002) Representing high-dimensional data sets as closed surfaces.
Information Visualization 1: 111-119.