Posts: 160
Joined: Wed Sep 22, 2010 12:26 pm


Post by glegrady » Thu Jan 13, 2011 11:47 pm

Please post your preliminary concept definition and doodles for the first project here. The preliminary concept and doodles are due on Tuesday, January 18 before class so that we can jump right in at 11am.
George Legrady

User avatar
Posts: 5
Joined: Wed Jan 05, 2011 10:34 am


Post by sandeepkbhat » Sun Jan 16, 2011 3:21 pm

Transactions as a function of Dewey numbers and the first character of the title.

In our own UCSB library, items are organized by call numbers which are derived based on subject of the item, the last name of the first author and year in which the item was released. The subject decides which section of the library the item lives in and the last name of the author decides which "rack" it would live in. So one can model the number of racks that are empty or full in a section to be dependent on the transactions modeled as a function of starting characters of the authors' names. I am curious to know if there are any interesting patterns to be found by observing the transactions based on subject (in SPL database that would be Dewey numbers) and author names. Since the SPL database doesn't include author names, I have decided to work with starting letter of the titles.

Initial thoughts on the project:
[x] The final graph would look similar to Catalog Tree's "Diplomatic Parking Violations" plot that we saw in class (sorry if my doodle doesn't convey this entirely).
[x] The X-axis would contain the letter of the alphabet and numbers, that is the first character of the title. I plan to exclude words like "The", "A", "An", "In", "On", and "Of" while doing the counting. I included numbers as I expect to see titles like "2012". There is an Others category to handle non-alpanumeric characters and possible characters from non-english languages.
[x] The Y-axis would contain the Dewey numbers; sections that have sub classes of my interest would be divided finer than the rest, for e.g. between 700 and 800, 775 indicates Digital Photography items that I have interest in and 776 includes Computer Art.
[x] Each point in the plot would be a boxed colored pixel. I plan to have colors going from blue to red, sort of a heat-map (blue being cold or low transaction regions and red being hot or high transaction regions). If that doesn't look good then I might consider using the same color map as that used in Catalog Tree's "Diplomatic Parking Violations" plot
[x] At the end of each row and column I plan to indicate the average number of transactions for that row or column.
[x] To begin with I want to do this for one year, say 2005. Time permitting I want to re-use the above plotting code to get plots for other years and maybe even put them together into animation showing the evolution over time.
Sandeep Bhat doodle for project 1

Posts: 1
Joined: Tue Jan 11, 2011 11:03 am


Post by Prud » Mon Jan 17, 2011 10:18 am

Hot Topics 2005-2010

My idea for the first project is to illustrate the relationship between library checkouts and real world events. There will be a line graph of the checkouts for each of the topics. Hopefully we will see significant changes around some important dates (e.g. Obama election, financial crisis, global warming summit, ..). Those topics at the doodle are not final, yet. I just picked some that could result in an interesting visualization.

The X-Axis will feature the time scale. I currently think about daily or weekly steps for each data point.
The Y-Axis will feature the number of items checked out (mapped to the total number of items checked out on that day/week) for each topic.

Additionally (for easier understanding) I might add a mark to the dates of important real world events.

The idea for the layout was influenced by the New York Times Michael Jackson Comparison example shown in the class.
Project 1 doodle (Patrick Rudolph)

Posts: 10
Joined: Mon Mar 29, 2010 1:33 pm


Post by dannyfan » Mon Jan 17, 2011 9:18 pm

M259 Visualizing Information | 2011 Winter | YYFan
PHASOR | 2-D Frequency Mapping

PHASOR is inspired by "2-D frequency" mapping. The definition of phasor from wikipedia is "In physics and engineering, a phase vector ("phasor") is a representation of a sine wave whose amplitude (A), phase (θ), and frequency (ω) are time-invariant."

"Euler's formula indicates that sine waves can be represented mathematically as the sum of two complex-valued functions..., it is understood to be a shorthand notation, encoding the amplitude and phase of an underlying sinusoid..., The sine wave can be understood as the projection onto the real axis of a rotating vector on the complex plane. The modulus of this vector is the amplitude of the oscillations, while its argument is the total phase ωt + θ. The phase constant θ represents the angle that the complex vector forms with the real axis at t=0."
The above concept also happens to be another inspiration for later animated visualization use.

PHASOR is therefore both visual metaphor and technique for visualizing frequency information in this project. Keywords, title, dewey class, check out time, and other statistics in database of Seattle Public Library are of interest for dimension of visual representation. Use of color in visualizing correlation between datasets is also of interest.

Posts: 2
Joined: Tue Apr 06, 2010 9:12 am


Post by qian » Mon Jan 17, 2011 11:33 pm

MAT 259 | 2011W | Qian Liu
What is Seattle famous for? Amazon, Starbucks and preforming Arts.
Seattle owns numerous numbers of of famous Internet, technology and software companies. ... in_Seattle
Seattle owns the original Starbucks and is famous for coffee industry.
Seattle has been a regional center for the perfuming arts for many years.

I'm interested in given the data a "local" meaning rather than just treat it as general data. Since most of the people who is using the library should be local residents, I want to find out the relationship between the books being checked out and local residents' interest. Four identical topics have been chosen for this visualization: technology(software, internet and computer science), Arts(performing Arts), Food and Drink and Economy as they all have a closer relationship with the city -- seattle.

The X-Axis will feature the dewey number(topics.)
The Y-Axis will feature the amount of the book that being checked out.
The larger the region is, means the larger the number is. The reason I picked hexagons over circles is to simulate a more interesting visual. The center of each hexagon region should be the darkest and the color will gradually fade out.
MAT259 Project1 Doodle

Posts: 10
Joined: Fri Sep 24, 2010 1:26 pm


Post by Amusesmile » Tue Jan 18, 2011 12:04 am

I’m interested in how we stereotype different countries/cultures/people. There was a pretty entertaining web app called the “Google Auto Complete Venn Diagram Generator” where you could see how google would complete a phrase based on the popularity of past searches. If you input countries, something like, “Why is Italy so…” you’d come up with a list like:


Anyway with the Seattle Public Library information, I’d like to visualize two things:

1. What countries are most frequently referenced in either the title or subject keywords of circulated items.
2. For each of these countries, what other words are they most often associated with.

I think this could reveal interesting patterns, especially if we compared this to different libraries or Google searches, etc. So far, I’ve made a preliminary sketch in Processing, which is basically making the visualization just with a collection of random words as opposed to the actual mySQL data. It looks sort of like a Wordle cloud.
Screen shot 2011-01-18 at 12.01.04 AM.png
Screen shot 2011-01-17 at 9.20.06 PM.png
Screen shot 2011-01-17 at 11.50.49 PM.png

Posts: 1
Joined: Thu Jan 13, 2011 11:07 am

Superhero popularity

Post by domagoj » Tue Jan 18, 2011 3:15 am

The most popular category in the dataset are comics and graphic novels, the goal of this interactive visualization is to enable exploration of the data related to superhero comics.

The menu on the left side lets the user select the comics for which data is displayed. Only comics that have had movies released between May 2005 and January 2011 are included in this visualization. All lines are color coded to the superhero, if possible the colors should be derived from the heroes costume.

The number of checkins are shown as a line, with a data point for every month between 2005 and 2011. Holding the mouse over a point on the line lists the names of the items that were checked in that month for that comic.

The time line has several ticks that mark dates when a superhero movie was released, holding the mouse over those ticks displays the name of the movie that was released and draws a vertical line over the checkin graph. This should provide an easy way to observe the effect (if any) of superhero movie releases on superhero comic checkins.

Posts: 4
Joined: Tue Jan 11, 2011 11:06 am


Post by nichole » Tue Jan 18, 2011 9:46 am

Checkout Durations
I am interested in the length of time items are checked out and how this varies across time and subjects:

- Time -
Are we checking out books for shorter and shorter durations? Or are we more forgetful about returning books on time? Or perhaps, books checked out on Thursdays are more likely to be kept longer? These are the driving questions behind the time plots I plan to create.
Attached is a time plot drawing. The number of days an item is checked are are plotted according to the day of the week that the item was checked. It is an interactive graph and scrolling over the bar for a specific day will highlight the other bars for that day and connect them by connecting the data points for that day. Also, there is a drop-down menu allowing users to view entire year averages or averages for specific weeks throughout the year.
- Subjects -
Do we see an increase in checkout duration times in response to world events? Do certain genres or categories of books tend to be checked out for longer periods of time? These are the questions I plan to explore with subject plots.
The second attached picture is my sketch of a subject plot. I have not yet decided if it will be clearer to plot the actual data points or simply plot points according to which subjects have the longest versus shortest durations. This graph will also allow the user to choose whether s/he views the overall averages or averages for specific weeks. However, I wanted to try to animate the graph with a "Play" button that will display the data for each week sequentially. Also, there is a "Timeline" button that will allow the user to look at the data over all the weeks at once in order to get an overview before zooming back in to a specific week.

Posts: 5
Joined: Mon Apr 12, 2010 10:16 am


Post by Solen » Tue Jan 18, 2011 9:47 am

Some of my ideas for the upcoming projects are:

1. Is there any way that the data (the contained) is related to the architecture (the container) of this specific building. Eventually, I may possibly create an abstract 3D model of the building with the way media is organized within it, and certain parts of it may glow intenser based on the usage/checkout at the time of the day

2. Does some possible peak moments in the recent news (i.e. market crash, the rise of foreclosures, the hurricane, the war, etc) change the reading/media trends related to these. If yes, what is the latency between the actual events, and people getting curious and reading about these. Or else, does it not effect the reading trends at all? Or do people deal with it in a more escapist behavior, such as indulging themselves more with fiction?

3. The time interval that popular media/book is released, and makes its first checkout in the library. Comparison of these intervals...

4. The correlation between the checkout period/interval and the actual average weight of the media type...
Time vs. weight. Print vs. electronic media: i.e. books are heavier than CD's and are tended to be checked-out for longer intervals.

5. Media landscape. Creating 3D temporal topographies representing the checkout averages of different media. Print, film, tape, CD, DVD, e-book, etc, all appear as different hills their sizes and height based on their check-out rate, and these landscapes/topographies are changing over time, shrinking, or growing, etc...

For this particular assignment (1st assignment: freq. mapping), I am taking a much simpler approach to familiarize myself with the data and visualizing approaches. Please see the attachment for this assignment.

Posts: 6
Joined: Tue Jan 18, 2011 9:23 am


Post by yerkes » Tue Jan 18, 2011 10:24 am

also, i'd like to use Code Page 437 graphics, maybe.
Codepage-437.png (2.45 KiB) Viewed 6455 times
Last edited by yerkes on Tue Jan 18, 2011 12:19 pm, edited 1 time in total.

Post Reply