Explode to Explain

From InfoVis:Wiki
Jump to navigation Jump to search

Authors


Short description

The basic problem behind this technique is, the complexity and sophistication of visualization techniques used to visualize constantly growing large data volumes and complex interrelationships. Traditional and easily self-explanatory techniques are not appropriate to visualize more complex interrelationships. On the other hand more complex visualization techniques are often very hard to understand for non expert people.

Exploded views are used widely in for example technical drawings. The idea is to show all details of a complicated construction by exploding the components and then showing the relationships of the components. Luboschik and Schumann (2007) propose the use of explosion for explaining the more complex information visualization technologies.

The explosition of basic visualization can be summarized in four steps.

1. Select a region of interest in the visual representation that will be used as akey element to explain the technique.
2. Generate an appropiate explosion scheme based on the underlying techniques.
3. Explode in comprehensive way.
4. Superimpose explanatory elements.

Explosion is used to represent the visualization techniques like data representations and structure visualizaion in a way that non-expert people can also understand the techniques easily. The example techniques for this are pixelmaps (Figure 1.) and treemaps (Figure 2.). However the same idea can be used to generate enhanced views af data (Figure 3.).


Figures


Figure 1. Exploded view of stepwise separation of recursion pattern.

In Figure 1. is explained the recursive pattern technique by using the explosion. All the steps of separation are shown from left to right and additional annotations are added to explain the semantic of recursion. One recursive pattern represents the data of one year, that is separated into its quarters, months and so on.



Figure 2. Exploded explanation of a treemap

Treemap is exploded to a serie of pictures. It shows the view-dependent explosion intodepth and the superimposing of underlying hierarchy. This kind of ilustration is applicable to most implicit 2D-techniques.



Figure 3. Highlighting the important data items by adding exploded view to a) recursive pattern technique, b) treemap

In recursive pattern the five highest values are pointed by arrows (left) and the data of those points is shown in right (Figure 3. (a)). In treemap the three most complex biological orders within the phylum of vertebrates are accentuated and labeled with their Latin name and their complexity value (Horton-Strahler number) (Figure 3. (b)).


Suitable Techniques / Datatypes

Basically all kind of implicit 2D-techniques and even some 3D- techniques, pixel oriented techniques. Also suitable for data in reprensented with techniques like described before.

Important Citations

.....up to now exploded views are not sufficiently considered in information visualization, although they are an integral part in other application scenarios. Therefore, in the following sections we demonstrate by the use of examples, that exploded views can also be useful in the context of information visualization to explain techniques and data.
[Luboschik & Schumann, 2007]


...we want to support the application of complex techniques by explaining them. For this reason, we concentrate on presentation rather than exploration and suggest a temporal adaptation of visual representations by an explosion and the integration of explanatory elements: we propose the illustration of images gained from information visualization.
[Luboschik & Schumann, 2007]


Explode to explain is not designed to replace any complex visualization technique, but to enhance them.
[Luboschik & Schumann, 2007]


Evaluation

The use of exploded views needs mote screen space than non exploded view. However by choosing the region of interest well, the needed space can be minimized.

This work is developed in a interdicipline environment and the non-expert people have told the exploded views really help them understanding the complex visualization techniques. A real user study is under development to get qualitative feedback of this approach.

Internal Links

References

Luboschik M. & Schumann H. 2004. Explode to Explain – Illustrative Information Visualization, Institute of Computer Science, University of Rostock, D-18051 Rostock, Germany, 11th International Conference Information Visualization

Evaluation References

Luboschik M. & Schumann H. 2004. Explode to Explain – Illustrative Information Visualization, Institute of Computer Science, University of Rostock, D-18051 Rostock, Germany, 11th International Conference Information Visualization