Advanced Visualization Techniques for Self-organizing Maps with Graph-Based Methods

The Self-Organizing Map is a popular neural network model for data analysis, for which a wide variety of visualization techniques exists. We present a novel technique that takes the density of the data into account. Our method defines graphs resulting from nearest neighbor- and radius-based distance calculations in data space and shows projections of these graph structures on the map. It can then be observed how relations between the data are preserved by the projection, yielding interesting insights into the topology of the mapping, and helping to identify outliers as well as dense regions.

• Type: Conference
• Year: 2005
• Where: ISNN
• Authors: Georg Pölzlbauer, Andreas Rauber, Michael Dittenbach
• Keywords: ISNN 2005, Method Defines Graphs, Popular Neural Network, Radius-based Distance Calculations