Conceptualizing Visual Uncertainty in Parallel Coordinates

Abstract

Uncertainty is an intrinsic part of any visual representation in visualization, no matter how precise the input data. Existing research on uncertainty in visualization mainly focuses on depicting data-space uncertainty in a visual form. Uncertainty is thus often seen as a problem to deal with, in the data, and something to be avoided if possible. In this paper, we highlight the need for analyzing visual uncertainty in order to design more effective visual representations. We study various forms of uncertainty in the visual representation of parallel coordinates and propose a taxonomy for categorizing them. By building a taxonomy, we aim to identify different sources of uncertainty in the screen space and relate them to different effects of uncertainty upon the user. We examine the literature on parallel coordinates and apply our taxonomy to categorize various techniques for reducing uncertainty. In addition, we consider uncertainty from a different perspective by identifying cases where increasing certain forms of uncertainty may even be useful, with respect to task, data type and analysis scenario. This work suggests that uncertainty is a feature that can be both useful and problematic in visualization, and it is beneficial to augment an information visualization pipeline with a facility for visual uncertainty analysis.

What May Visualization Processes Optimize?

“…4. This uncertainty can be caused by inaccuracy or aggressive aggregation of a machine-centric transformation, as well as by human factors such as visual uncertainty [19] and lack of understanding and experience.…”

“…Assuming that we represent share price at USD $0.01 resolution using 32-bit unsigned integers, the minimum and maximum values are thus 0 and 2 32 − 1 cents respectively. (Note: Historically the highest share price in the US is 347,600 cents, i.e., 2 18 < 347, 600 < 2 19 ). If the probability of different time series were uniformly distributed, the entropy of this alphabet would be 23040 = 720 × log 2 (2 32 ) bits.…”

The importance of tracing data through the visualization pipeline

“…Is there data hidden behind other data? [9] Often, the user may not even be aware that data can be hidden, much less have a way of finding out whether there is hidden data or not.…”

“…In previous work we have presented a taxonomy of the various sources, causes and effects of visual uncertainty in parallel coordinates [9].…”

“…We achieve this by extending the Chi model [7] of the visualization pipeline ( Figure 1) by including the perception and cognition stages: data mapping and visual mapping constitute the encoding stages on the machine-side, while perception and cognition constitute the decoding stages on the human side of the pipeline. Then we add a feedback loop that is informed about visual uncertainty [9] and can be used to measure visual representations according to he different levels of uncertainty in the screen-space.…”