By Daniel Perry, PhD student
This past February, I presented at the 2013 iConference in Fort Worth, Texas. The theme of this year's conference was "Data, Innovation, and Wisdom," with a wide range of topics that included information system design, data visualization, social media analysis, and even gaming research. I presented a paper, "VizDeck: Streamlining exploratory visual analytics of scientific data," co-authored with my advisor in Human Centered Design & Engineering (HCDE), Professor Cecilia Aragon, as well as with Computer Science & Engineering (CSE) Professor Bill Howe and former CSE Research Engineer Alicia Key. My presentation described the design and evaluation of VizDeck, a web-based visual analytics tool that automatically recommends a set of appropriate visualizations based on the statistical properties of the data and adopts a card game metaphor to present the results to the user. The talk was one of three papers presented in the visualization session this year.
I attended the big social media data workshop at the start of the conference, which centered on the challenges of collecting and analyzing large social media datasets. Much of the focus of the workshop was on methods and tools for capturing and analyzing data on Twitter keywords and hashtags, with Sean Goggins and Alan Black presenting research out of the Group Informatics Lab at Drexel University. There was an interesting discussion among those at the workshop on appropriate methods for sampling (e.g., which Tweets should be selected for further analysis) as well as qualitative methods for analysis (such as a grounded theory approach to coding the content of Tweets). Lively discussions ensued as workshop goers noted the challenges of analyzing social media data that was much less structured than what was afforded by Twitter's API, such as data from online forums. The lack of existing tools for analysis of less structured social media data is a pain point that my lab, the Scientific Collaboration and Creativity (SCC) Lab, is working on addressing with in our creation of TextPrizm.
There were numerous talks this year that integrated social media topics, such as Kimberly Glasgow's (University of Maryland) presentation on the social media response of public organizations during the 2011 London riots, and Toine Bogers and Lennary Bjorneborn's (Royal School of Library and Information Science) interesting analysis of #serendipity on Twitter. Some of the session topics melded newer genres with more classical information science theories. I thoroughly enjoyed Julia Bullard's (University of Texas) presentation on how player generated tutorials for World of Warcraft offer important insights for integrating playfulness into information resources.
This being the second iConference I've attended, and it's been exciting to watch the conference grow as it branches out from being primarily a meeting of the Information Schools to becoming a more expansive venue for those interested in information science from a variety of perspectives, including computer science, information and library science, humanities, and human-computer interaction (HCI). I hope to be able to attend next year in Berlin.