Visual analytics supports humans in generating knowledge from large and often complex datasets. Evidence is collected, collated and cross-linked with our existing knowledge. In the process, a myriad of analytical and visualisation techniques are employed to generate a visual representation of the data. These often introduce their own uncertainties, in addition to the ones inherent in the data, and these propagated and compounded uncertainties can result in impaired decision making. The user's confidence or trust in the results depends on the extent of user's awareness of the underlying uncertainties generated on the system side. This paper unpacks the uncertainties that propagate through visual analytics systems, illustrates how human's perceptual and cognitive biases influence the user's awareness of such uncertainties, and how this affects the user's trust building. The knowledge generation model for visual analytics is used to provide a terminology and framework to discuss the consequences of these aspects in knowledge construction and though examples, machine uncertainty is compared to human trust measures with provenance. Furthermore, guidelines for the design of uncertainty-aware systems are presented that can aid the user in better decision making.

Visual analytics enables us to analyze huge information spaces in order to support complex decision making and data exploration. Humans play a central role in generating knowledge from the snippets of evidence emerging from visual data analysis. Although prior research provides frameworks that generalize this process, their scope is often narrowly focused so they do not encompass different perspectives at different levels. This paper proposes a knowledge generation model for visual analytics that ties together these diverse frameworks, yet retains previously developed models (e.g., KDD process) to describe individual segments of the overall visual analytic processes. To test its utility, a real world visual analytics system is compared against the model, demonstrating that the knowledge generation process model provides a useful guideline when developing and evaluating such systems. The model is used to effectively compare different data analysis systems. Furthermore, the model provides a common language and description of visual analytic processes, which can be used for communication between researchers. At the end, our model reflects areas of research that future researchers can embark on.

Information visualisation is about gaining insight into data through a visual representation. This data is often multivariate and increasingly, the datasets are very large. To help us explore all this data, numerous visualisation applications, both commercial and research prototypes, have been designed using a variety of techniques and algorithms. Whether they are dedicated to geo-spatial data or skewed hierarchical data, most of the visualisations need to adopt strategies for dealing with overcrowded displays, brought about by too much data to fit in too small a display space. This paper analyses a large number of these clutter reduction methods, classifying them both in terms of how they deal with clutter reduction and more importantly, in terms of the benefits and losses. The aim of the resulting taxonomy is to act as a guide to match techniques to problems where different criteria may have different importance, and more importantly as a means to critique and hence develop existing and new techniques.

Previous work has demonstrated the use of random sampling in visualising large data sets and the practicality of a sampling lens in enabling focus+context viewing. Autosampling was proposed as a mechanism to maintain constant density within the lens without user intervention. However, this requires rapid calculation of density or clutter. This paper defines clutter in terms of the occlusion of plotted points and evaluates three possible occlusion metrics that can be used with parallel coordinate plots. An empirical study showed the relationship between these metrics was independent of location and could be explained with a surprisingly simple probabilistic model.

We have previously shown that random sampling is an effective clutter reduction technique and that a sampling lens can facilitate focus+context viewing of particular regions. This demands an efficient method of estimating the overlap or occlusion of large numbers of intersecting lines in order to automatically adjust the sampling rate within the lens. This paper proposes several ways for measuring occlusion in parallel coordinate plots. An empirical study into the accuracy and efficiency of the occlusion measures show that a probabilistic approach combined with a 'binning' technique is very fast and yet approaches the accuracy of the more expensive 'true' complete measurement.

This paper describes two visualisation algorithms that give an impression of current activity on a web site. Both focus on giving a sense of the trail of individual visitors within the web space and showing their navigation paths. Past web activity is used to produce a spatial mapping of pages, which results in highly traversed page links lying close together in the 2D visualisation space. Pages visited by typical individual visitors thus form intelligible paths when plotted in the visualisation space. Both techniques attempt to enhance user awareness and experience, but they differ in their balance between utility and aesthetics.

This paper presents a technique, the Quantum Web Field, designed to give an ambient visualisation of the current activity on a Web site. It uses the paths of past visitors to the site and a self-organising map to build a diffuse 'probabilistic' mapping of pages to cells in a 2D matrix, where highly traversed page-links tend to be closer to each other. The paths of current visitors appear as intelligible trails giving a sense of purposeful human activity rather than offering detailed analysis. The visualisation is not constrained by either the complexity or the number of pages in the site.

The use of random algorithms in many areas of computer science has enabled the solution of otherwise intractable problems. In this paper we propose that random sampling can make the visualisation of large datasets both more computationally efficient and more perceptually effective. We review the explicit uses of randomness and the related deterministic techniques in the visualisation literature. We then discuss how sampling can augment existing systems. Furthermore, we demonstrate a novel 2D zooming interface - the Astral Telescope Visualiser, a visualisation suggested and enabled by sampling. We conclude by considering some general usability and technical issues raised by sampling-based visualisation.

Information visualisation systems often have to cope with presenting large amounts of data. In this paper we look at the problem of reducing the display density through the use of random sampling. We review some of the uses of sampling and examine density reduction techniques currently used in visualisation. We then look at a proposed novel visualisation that utilises sampling and discuss many issues that arise. Furthermore, we propose the /spl zeta/DB sampling database architecture and demonstrate how this can be used to enhance current visualisation applications. Finally, we describe the architecture of a sampling aware application and illustrate its advantages.

Interactive visualisation has been one of the most exciting areas in HCI over recent years. The key term here is 'interactive', and in this paper we assert that virtually any static representation can become more powerful by the addition of simple interactive elements. This is demonstrated by adding interactivity to standard representations including stacked histograms, pie charts and scatter plots. We show how adding interactivity can help resolve many of the trade-offs inherent in static visualisations by allowing multiple options to be available and most importantly for them to be interactively related. Many years of creativity and effort have been invested in traditional generic and bespoke visualisations. Adding interactivity leverages this accumulated experience, but also adds an extra dimension.

No abstract available.