The Community for Technology Leaders
Green Image
Issue No. 06 - November/December (2010 vol. 16)
ISSN: 1077-2626
pp: 980-989
David Feng , UNC Chapel Hill
Lester Kwock , UNC Chapel Hill
Yueh Lee , UNC Chapel Hill
Russell Taylor , UNC Chapel Hill
Conveying data uncertainty in visualizations is crucial for preventing viewers from drawing conclusions based on untrustworthy data points. This paper proposes a methodology for efficiently generating density plots of uncertain multivariate data sets that draws viewers to preattentively identify values of high certainty while not calling attention to uncertain values. We demonstrate how to augment scatter plots and parallel coordinates plots to incorporate statistically modeled uncertainty and show how to integrate them with existing multivariate analysis techniques, including outlier detection and interactive brushing. Computing high quality density plots can be expensive for large data sets, so we also describe a probabilistic plotting technique that summarizes the data without requiring explicit density plot computation. These techniques have been useful for identifying brain tumors in multivariate magnetic resonance spectroscopy data and we describe how to extend them to visualize ensemble data sets.
Uncertainty visualization, brushing, scatter plots, parallel coordinates, multivariate data

L. Kwock, R. Taylor, Y. Lee and D. Feng, "Matching Visual Saliency to Confidence in Plots of Uncertain Data," in IEEE Transactions on Visualization & Computer Graphics, vol. 16, no. , pp. 980-989, 2010.
95 ms
(Ver 3.3 (11022016))