An application of the Self-Organizing Map and interactive 3-D visualization to geospatial data

Computer technologies have been rapidly improving throughout the last couple of decades, and they are now at the stage of allowing scientists to carry out data analyses that deal with very complex and multivariate datasets. Moreover, there are growing numbers of researchers who wish to carry out such tasks in real-time. Traditional data analyses and visualization techniques are useful but not sufficient to achieve those tasks. The Self-Organizing Map (or Kohonen’s Feature Map) is one of the many modern data analysis tools that researchers have found useful in analyzing high-dimensional (multivariate) datasets such as atmospherical and demographical data. It is often used for such data analyses because of is multidimensional scaling and topological mapping capabilities. However, information loss caused by multidimensional scaling sometimes results in difficulty in interpreting an SOM when it is visualized in 2-D space. This study presents the use of the SOM for geospatial data analysis with the help of Java-based advanced 3-D visualization tools and a visual programming environment (GeoVISTA Studio) in order to gain deeper understanding of those complex datasets.