New tool for clustering and analyzing spatial data with Neural Networks and Neural Gas

A new tool for clustering and analyzing geographic data with artifical self-organizing neural networks (SOM) and the innovative Neural Gas (NG) algorithms has been made availabe. The free SPAWNN suite supports different spatial context models and it also establishes interactive linkage between the neural network and geographic maps.
Notably it enables further the follow-up clustering of the resulting network models!
The tool is available at http://koenigstuhl.geog.uni-heidelberg.de/spawnn/ along with some sample data. The concept of contextual neual gas for spatial clustering and analysis is being introduced in this publication:
Hagenauer, J. and Helbich, M. (2013). Contextual neural gas for spatial clustering and analysis. International Journal of Geographical Information Science, 27(2):251–266. See here for some further examples for spatial analysis with Neural networks and in particular SOM and extensions thereof.