Numbers, Words and Colors
Description: Tools developed by Martin Wattenberg and his associate Fernanda Vi_gas, have changed the way people look at and use visualizations, by empowering and equipping users with the methodology needed to ask different questions. Wattenberg, whose background is in math and computer science, asks how the humanities have influenced the evolution of data visualization and then answers with several examples from his own work.
Web Seer compares Google's "auto"suggest" feature in one"to"one, weighted comparisons such as "why doesn't he" and "why doesn't she" The resultant text image uses the size of arrows and words to reflect frequency, demonstrating how text can impart meaning.
Another Wattenberg/Vi_gas collaboration is Many Eyes, a social media tool and Web site that has "democratized" powerful visualization systems by putting them in the hands of general audiences. This tool lets users visualize data in numerous ways, from scatterplots and bar charts to tree maps and stack graphs.
Word Tree, a visualization technique that lets users pick a word or phrase from a data set, shows the different contexts in which it appears via a tree"like branching structure. Chimera takes care of the "boilerplate problem" by examining large collections of text, such as contracts, and pointing out identical phrases. Seeing results arranged in faux 3D "skyscrapers" clearly illustrates levels of recurrence. Although Word Tree and Chimera are fundamentally repetition searches, they are important tools for semantic analysis: simple, but revealing.
The idea behind Phrase Net is to expose a text's underlying network; this visualization tool diagrams the relationships between different words used in a text. It uses a simple form of pattern matching to provide multiple views of the concepts contained a book, speech, or poem.
Another Wattenberg/ Vi_gas collaboration is Fleshmap, "an inquiry into human desire." The relationship between the body and its visual and verbal representation are explored in a series of artistic studies employing song lyrics and body imagery. Flickr Flow, Wattenberg explains, is an experiment whose materials are color and time. Software calculated the relative proportions of different colors seen in photos of Boston taken during each month of the year and plotted those colors on a wheel creating a "river of meaning."
Wattenberg addresses questions regarding the impact of race in personified visualizations, and his subjective motives in selecting particular data for analysis. He admits that his "hard drive is loaded with failed visualizations," but emphasizes that the visualization process should be one of trial and error. As for encouraging the development of visual literacy, Wattenberg concludes, "as visualization becomes part of the discourse and people realize, 'this is something that's powerful, it can help me make my case in life,' they'll learn I'm hoping for education and good, old"fashioned human brain power."
About the Speaker(s): Martin Wattenberg is a computer scientist and artist. From 2005 to 2010, he founded and managed IBM's Visual Communication Lab, exploring new forms of visualization and how they can enable better collaboration. A key project, Many Eyes, is an experiment in open, public data visualization and analysis. Prior to joining IBM, Wattenberg was the Director of Research and Development at SmartMoney.com, a joint venture of Dow Jones and Hearst. His work at SmartMoney included the groundbreaking Map of the Market.
Wattenberg is also known for his visualization"based artwork, which has been exhibited in venues such as the London Institute of Contemporary Arts, the Whitney Museum of American Art, and the New York Museum of Modern Art.
Wattenberg earned a B.S. at Brown University (1991), M.S. at Stanford (1992) and a Ph.D. at the University of California Berkeley (1996).
Host(s): School of Humanities, Arts & Social Sciences, HyperStudio
It looks like no one has posted a comment yet. You can be the first!
More from MIT World — special events and lectures
Added over 2 years ago | 01:21:00 | 3584 views
Added almost 3 years ago | 01:33:00 | 3276 views
Added over 2 years ago | 00:47:19 | 3030 views
Added almost 3 years ago | 01:30:00 | 2059 views
Added almost 3 years ago | 00:45:34 | 1394 views
Added over 2 years ago | 01:25:00 | 2497 views