'Circle Segments': A Technique for Visually Exploring Large Multidimensional Data Sets
Mihael Ankerst, Daniel A. Keim, Hans‐Peter Kriegel · KOPS (University of Konstanz) · 1996
In this paper, we describe a novel technique for visualizing large amounts of high-dimensional data, called ‘circle segments’. The technique uses one colored pixel per data value and can therefore be classified as a pixel-per-value technique [Kei 96]. The basic idea of the ‘circle segments ’ visualization technique is to display the data dimensions as segments of a circle. If the data consists of k dimensions, the circle is partitioned into k segments, each representing one data dimension. Inside the segments, the data values belonging to one dimension are arranged from the center of the circle to the outside in a back and forth manner orthogonal to the line that halves the segment. Our first results show that the ‘circle segment’ technique is very powerful for visualizing large amounts of data, providing more expressive visualizations than other wellknown techniques such as the ‘recursive pattern ’ technique and traditional