Laboratory of Computer and Information Science > Teaching > T-61.5010 > Exercise work 2007

Exercise work 2007

NOTICE!!! As some you may have observed, the server that is hosting some of the data sets for Task 2 is not responding. The problem will hopefully be fixed soon. Because of this, the deadline for Task 2 is extended to next friday, that is, March 16. This deadline is final, and will not be further extended even if kdd.ics.uci.edu stays down. Task 1 should still be submitted by the original deadline of March 12, however. Those of you who have already submitted the report may hand in an updated version before the new deadline if you still want to improve your work.

To pass the course you must pass the examination and complete this exercise work. The work will be graded on the scale rejected, passed and passed with distinction.

To complete the exercise work you may need information that has not been discussed in the lectures. If you have any questions please do not hesitate to contact the assistant or the lecturer for hints, guidance or references.

The exercise work will be graded within a month of the deadline. A passed exercise work will be valid for one year after the original deadline.

General requirements

1. The exercise work should be completed by one person. However, discussing it with others is encouraged.
2. The deadline is March 12 2007. (Due to problems with the kdd.ics.uci.edu server the deadline for Task 2 is extended to March 16.)
3. Reports submitted after the final deadline will be rejected. If you have a very good reason that causes you to miss the final deadline you can request an extension. The extension must be requested well in advance before the deadline.
4. To pass the exercise work you must perform both tasks given in the "specific requirements" section.
5. To pass with distinction you have to carry out the requirements exceptionally well.
6. Accepted languages for the report are Finnish, Swedish and English.
7. If you submit the exercise work report in time, but don't pass, you will have a chance to supplement your work after the deadline. The supplemented project works will not be eligible for the higher grade, "passed with distinction".
8. Each exercise work report should contain a section that comments on the difficulty of the project and an estimate of the time used for completing it.
9. Documentation should be in a format readable with generally available tools. Preferred formats are Postscript and PDF. However, HTML and plain text are accepted as well.
10. The exact deliverables are indicated below, but just to emphasize: We do not expect you to hand in working applications or source code! This is not a programming exercise, even though completing it does probably involve some programming.
11. The exercise work reports should be submitted by email (t615010@james.hut.fi). You can either provide a WWW link from which your report can be downloaded or you can attach the report to the email. If email submission is not possible you can use e.g. (internal) mail (Antti Ukkonen, PL 5400, 02015 TKK). The exercise work reports should contain your name, email address and full student number. If you submit your work by email, please include the student number also on the subject line.

Specific requirements

Task 1

Write a summary (roughly 1000 words) of the paper The Value of Visualization, Jarke J. van Wijk, Infovis 2005.

The task is to produce a self-contained document that can be understood by anybody who has taken this course but hasn't read the actual paper by van Wijk. Concentrate on the important parts of the paper, do not try to discuss everything in the 1000 words! You can use images in the summary to clarify important concepts, but if you do, make sure they are referenced in the text and have a proper caption. You can also use images from the original article, but if you do so, please indicate the source clearly. And do not use images simply to make your summary look more interesting, they must work in conjunction with the text.

Task 2

Download one of the data sets listed below and visualize some interesting phenomena in the data. Write a short (3-5 pages) report that describes your approach, obtained results and the methods and tools used.

The purpose is not to do an analysis that covers the data from all possible angles. Focus on one or two aspects and try to make a clear visualization of them. In some cases the data descriptions mention tasks for which the data was originally published. Do not care about these! Try to relate your work to the topics discussed in the course, i.e., Tufte's principles, Gestalt laws, etc., when preparing the visualizations. Also, make sure your images convey at least some of their information when printed on a black and white laser printer.

NOTICE! (Added Sat. 10.3, 2.30 pm) The kdd.ics.uci.edu site seems to be unavailable. This is probably a temporary problem, but if you haven't yet obtained a copy of a data set hosted on the site in question, please use some other data that is available. Unfortunately we do not have copies of the data sets that we could host on the local server. At the moment the problem only affects the Internet Usage Data.

• Internet Usage Data
• The Insurance Company Benchmark If the previous link doesn't work, try this instead!
• US Census 1990 data
  The data consists of a random sample of 24999 individuals, each described by 68 attributes, such as level of education, occupation, country of origin, number of children, etc. More information can be found at the UCI KDD Archive website. Do not try to analyze the complete data. Instead, select some interesting subset of the attributes for inspection. Descriptions of the attributes can be found here. (If the previous link does not work, try this instead.) Note that the continuous variables (eg. age) have been discretized by using this script. For example, the value 5 in the dAge column means the person is 40-49 years old. Check the script for further details. It is written in SQL but should be very easy to interpret for anybody with programming experience.

The use of available visualization tools (e.g. Matlab, R, Inkscape, (Open)Office, etc.) is encouraged. However, implementing your own scripts and small programs for processing the data will be necessary. Do not hand in any programs or source code you may have written! They will not be graded.