(due Wednesday, March 14 – enough before class that I can look over the responses)
This reading is meant to give you a sense of some of the ways people try to present high-dimensional data.
You should (quickly) read this old paper:
- Wong and Begerton. 30 years of Multidimensional Multivariate Visualization (find it here). 1997
This paper is archaic – but I want you to see it to get a sense of the “zoo” of not-so-great ideas – many of which did not stand the test of time. Here’s another old survey, mainly useful as a gallery of methods.
- “High-Dimensional Visualizations,” G. Grinstein, M. Trutschl, U. Cvek, 7th Data Mining Conference-KDD 2001, San Francisco, California, 2001. [PDF]
You should look over both papers to get a sense of the kinds of methods that have been developed over the years.
Two particularly important techniques are scatterplot matrices and parallel coordinates displays. These are so common, there’s no real reading for them but…
For scatterplot matrices, check out the scatterdice system. (project page) rather than reading the paper, you can watch the video.
For parallel coordinates… I want you to find an example of them on the web. Preferably one that offers some degree of interaction (since that’s one of the key ways to address their issues). In your Piazza post, tell us where you found one.
If you’re curious about parallel coordinates, this paper provides a generalization to “non-parallel” coordinates
- Claessen JHT, van Wijk JJ. Flexible Linked Axes for multivariate data visualization. IEEE transactions on visualization and computer graphics. 2011;17(12):2310-6. Available at: http://www.ncbi.nlm.nih.gov/pubmed/22034351. Accessed March 9, 2012. (this reading is optional for class)
On the Piazza page, give your thoughts on the various methods you’ve seen.