Self-organization in parallel coordinates

Abstract

Parallel coordinates has shown itself to be a powerful method of exploring and visualizing multidimensional data. However, applying this method to large datasets often introduces clutter, resulting in reduced insight of the data under investigation. We present a new technique that combines the classical parallel coordinates plot with a synthesized dimension that uses topological proximity as an indicator of similarity. We resolve the issue of over-plotting and increase the utility of the widely-used parallel coordinates visualization. © 2013 Springer-Verlag Berlin Heidelberg.