Daniel A. Keim
University of Halle-Wittenberg
Germany
keim@informatik.uni-muenchen.de
Visual data mining techniques have proven to be of high value in exploratory data analysis and they also have a high potential for database mining. A number of data visualization techniques has been proposed in this context - including our pixel-oriented visualization techniques which try to represent as many data items as possible on the screen at the same time by mapping each data value to a pixel of the screen and arranging the pixels adequately. Although a number of visualization techniques has been proposed in the literature, little work has been done on evaluating and comparing the techniques.
In this talk, we try to evaluate our visual data mining techniques and to compare them to other well-known visualization techniques for multidimensional data: the parallel coordinate and stick figure visualization techniques. For the evaluation of visual data mining techniques, in the first place the perception of properties of the data counts, and only in the second place CPU times and the number of secondary storage accesses are important. In addition to testing the visualization techniques using real data, we developed a testing environment for database visualizations similar to the benchmark approach used for comparing the performance of databases. The testing environment allows the generation of test data sets with predefined data characteristics which are important for comparing the perceptual abilities of visual data mining techniques. In addition to the perceptual performance comparison, we also investigate the time performance of our visual data mining techniques analytically and experimentally.