This page presents a selection of the best student work done during the EagerEyes class (page in German) held in the spring of 2005 at Vienna University of Technology.

Students chose one of three different datasets, and designed a visualization for it. The datasets were: Titanic, Cars, and Mammals. The goal of the assignment was to design a visually pleasing and useful visualization. The students did not have to implement their designs, so that they would not limit themselves to things that could be programmed easily. We had initially planned to define a subset of the designs of every group for implementation, but had to drop that due to the large number of students (63, and in the beginning over 70), and the limited resources (we were both not employed at the university, and only paid a rather symbolic amount as external lecturers).

Titanic

This data set turned out to be the most popular of the three. It contains one record for each of the 2229 passengers and crew, giving their sex, age (child or adult), class (1, 2, 3, or crew), and whether they survived or not.

It is interesting to compare the different abstraction levels at which the people were displayed. The most direct display certainly is to show each person as one object. In both of the following examples, the size of the point shows if the person was a child or an adult, and the color encodes the sex; the position tells the user the survival status.

Nenning, Veigl (Video)

Dabrowski, Jakl, May

Counting individual points is tedious, so it is useful to abstract, and to show the numbers rather than the individual people. Especially the first example works like an information graphic, and is very self-explanatory. These three examples also show how the ship or parts of it can be used to create a compelling visual metaphor.

Brandejsky, Buturovic, Kilzer

Graf, Post

Emsenhuber

Providing different views of the data in the same image enables the user to find out new interesting relations that were not directly encoded in the image. The following two examples are not immediately easy to understand, but if one knows the key, all kinds of relations within the data can be found.

Cecch, Schauer, Scholz

Satzer, Treutler, Banfield-Mumb

Cars

Data about car models that were introduced from 1970 to 1982 are included in this dataset. For each car model, the name, manufacturer, originating country, mileage, etc. are given. Using a visual metaphor, students managed to put many pieces of information into a glyph that is almost self-explanatory. It is possible to read the engine power, mileage, brand, and originating country from these images, and to compare different car models. A tooltip lists the exact values.

The following two variations on parallel coordinates are less metaphorical, but still provide some simple but useful additional information, that many visualization systems still lack: labels and scales.

Haberl, Haidacher, Zirknitzer

Froschauer

Mammals

This data set consisted of the mammalia subtree of the tree of species used in the InfoVis Contest 2003. This data set is different from the above because it contains little information per entry (only two names), but is large and structured in a tree. The hyperbolic tree certainly made a big impression on students, as can be seen in the following two examples.

Fahrngruber, Fellner, Maschek

Kusternig, Witzmann

The most creative solution we got for any of the assignments certainly was the following. The students had not only put a lot of effort into producing a video, but had also come up with some very thoughful solutions to common interaction problems. To fully appreciate this video, make sure to turn on your speakers.

Chiu, Daxböck, Domig