Technical Report 94-11

TITLE:

Test Data Sets for Evaluating Data Visualization Techniques

DATE:

June 1994

AUTHORS:

Daniel A. Keim

keim@informatik.uni-muenchen.de

Institut für Informatik

Universität München

Leopoldstr. 11B

D-80802 München (Germany)

R. Daniel Bergeron

rdb@cs.unh.edu

Department of Computer Science

University of New Hampshire

Durham, NH 03824

USA

Ronald M. Pickett

pickett@cs.uml.edu

Institute for Visualization and Perception Research

University of Massachusetts Lowell

Lowell, MA 01854

USA

KEYWORDS:

Testing Data Visualizations, Generating Test Data, Visualizing Multidimensional and Multivariate Data, Perception of Visualizations

ABSTRACT:

In this paper we take a step toward addressing a pressing general problem in the development of data visualization systems - how to measure their effectiveness. The step we take is to define a model for specifying the generation of test data that can be employed for standardized and quantitative testing of a system's performance. These test data sets, in conjunction with appropriate testing procedures, can provide a basis for certifying the effectiveness of a visualization system and for conducting comparative studies to steer system development.