Media Arts and Technology

The assignment is to create a 2D spatial map where 3 sets of data are compared to each other and positioned in a x,y,z position. Z can represent depth, through brightness. Coloration or form deviation may be the way to compare the two.

The first step is to decide how your data will map itself within a 2D space. You may have time mapped horizontally (hours of the day) and media type mapped vertically, or the whole Dewey sequence from 0-999. Top layer may be 2010, bottom layer may be 2011. Transparency would allow us to see differences.

Another approach may be to compare Dewey in the top layer, and non-Dewey in the bottom layer. Or the comparison could be the same dataset contrasted according some other aspect of the metadata we have: collcode, item type, etc.

This project will require some thinking of what could be compared, followed by SQL queries to see if there are interesting results, and then to make visualization tests.

To stimulate thinking Scientific Visualization PhD candidate Basak Alper will come and present her research on February 5. In preparation for this, please read her article "Weighted Graph Comparison Techniques for Brain Connectivity Analysis".

For visual stimulus, the Visual Complexity website has many examples of both node-connected data and matrix mapping: http://www.visualcomplexity.com/vc/

http://www.mat.ucsb.edu/basak.alper/

Hi guys,

I changed my idea, I used the code that Yoon wrote to show the ranks of the dewey classes in each hour and modified it so that you can look across each row (dewey category) and see how it's rank changes based on the brightness of the squares. You can see for example that dewey class 789 falls in rank at 8 and 9 am and also see how much the ranks fluctuate for each class based on the diference in brightness.

Also you can look at the vertical columns to see how the categories compare in rank to each other.
Screenshot and code attached.

My doodle is here.

https://docs.google.com/document/d/1LdT ... oyiaa8/pub

And my final code(sketch) is attached below.

thanks!

Here is my doodle link.

https://docs.google.com/file/d/0BxTPhw- ... YtVEU/edit


Yeu-Shuan Tang

Link to my Project 3:

https://docs.google.com/document/d/1lmL ... sp=sharing

The following visualization is an interactive ranking of the top six food cuisines at the Seattle Public Library by month. For example, the first month desert was the most popular food item followed by Spanish, Chinese, French, Vegetarian, and then Italian food.

If you hover over a rectangle it displays the rank for that food cuisine over the 5 years so you can more easily see the changes in ranking over the months.

Here is mine:
https://docs.google.com/document/d/16Ct ... sp=sharing

Attached is the zip file. Writeup is coming soon.

Sketch:
https://docs.google.com/drawings/d/1y7m ... X99LE/edit

Intro

I was curious to see how the current most popular dewey categories ranked 6 years back.

Query


select
year(activity.o) as 'Year',
sum(case when (dewey.dewey = 811) then 1 else 0 end) as '811 Poetry',
sum(case when (dewey.dewey = 599) then 1 else 0 end) as '599 Food & Drink',
sum(case when (dewey.dewey = 746) then 1 else 0 end) as '746 Textile Arts',
sum(case when (dewey.dewey = 398) then 1 else 0 end) as '398 Folklore',
sum(case when (dewey.dewey = 613) then 1 else 0 end) as '613 Health',
sum(case when (dewey.dewey = 616) then 1 else 0 end) as '616 Diseases',
sum(case when (dewey.dewey = 796) then 1 else 0 end) as '796 Sports',
sum(case when (dewey.dewey = 917) then 1 else 0 end) as '917 North America',
from
dewey, activity
where
dewey.bib = activity.bib and
dewey.dewey in (811, 599, 746, 398, 613, 616, 796, 917, 5) and
year(activity.o) in (2006, 2012)
group by year(activity.o);

Query Explanation

Each of the chosen dewey categories is queried, and the checkout occurrences are tallied with the 'sum' SQL function, then grouped by year.

Results and Analysis

Since I wanted to visualize the counts from both years simultaneously, and both years had data for the same categories, I thought a 'web graph' was perfect. These graphs are typically used for displaying personality traits, each axis being a different personality. Here each axis represents one of the preselected dewey categories, and by making the polygonal shape that stretches across the axes translucent, both years are visible. I considered including data from the years in between, but this method broke down with any additional translucent layers as the colors could no longer be differentiated from the graph.

Mostly a sketch at the moment: http://www.uweb.ucsb.edu/~andrewwaranis ... project_3/ .
Once I straighten this out, I can code the backend and compose my query.

Download: http://www.uweb.ucsb.edu/~andrewwaranis ... /dw_2d.zip .

[edit]

I finally got my laptop back working...
Trying to improve the previous projects
upgraded version



Project description
https://docs.google.com/document/d/1672 ... qmH6s/edit#


Sketch

A combination of my proposed sketch (sorry it's late) and my initial visualization:
https://docs.google.com/document/d/1Fr- ... PyMqs/edit

sketch
https://drive.google.com/?authuser=0#fo ... Wd4WUhqdjg

images:
https://docs.google.com/file/d/0B7PkP8N ... hZSGM/edit
https://docs.google.com/file/d/0B7PkP8N ... Fzdkk/edit
https://docs.google.com/file/d/0B7PkP8N ... t4Z1E/edit