How 'Image-Texts' May Reveal Social Trends: Lev Manovich to Present Findings

Professor Lev Manovich (Computer Science) has coauthored new findings on social computing to be presented at the IEEE 2015 Big Data conference (October 29-November 1).

Co-authored with Mehrdad Yazdani, the paper describes how the frequency of "image-texts" - such as screen shots of text chats or other non-photographic images - correlate with well-being responses from Gallup surveys, and also median housing prices, incomes, and education levels.

The paper is the first ever to analyze qualitatively large numbers of Tweeted images.

Data comes directly from Twitter as part of a Twitter Data grant awarded in 2014 to Manovich and his collaborators, who studied one million images shared on Twitter in 20 U.S. cities during one year.
 
For more details about the paper, titled "Predicting Social Trends from Non-photographic Images on Twitter," see the GC's Software Studies Initiative.

A separate project directed by Manovich, "On Broadway," is currently on view at the New York Public Library as part of its exhibition "Public Eye: 175 Years of Sharing Photography."