Two UMSI PhD students to present papers at computational science conference

Sam Carton

Two UMSI PhD students, Sam Carton and Gaurav Paruthi, will present papers at the ninth annual International AAAI Conference on Web and Social Media (ICWSM), held May 26-29 in Oxford, England.

ICWSM is one of the top venues for computational social science, serving as a forum for researchers in computer and social science to come together to share knowledge, discuss ideas, exchange information, and learn about cutting-edge research in diverse fields with the common theme of online social media.

Carton’s paper, “Audience Analysis for Competing Memes in Social Media,” was co-authored by UMSI faculty members Eytan Adar, Qiaozhu Mei, and Paul Resnick, as well as BSI student Nicole Zeffer and Souneil Park of Telefonica Research.

The researchers on the project developed a tool for audience analysis, focusing on the people who are passively exposed to messages, with a special emphasis on competing memes such as propagations and corrections of a rumor.

In such competing meme diffusions, important questions include which meme reached a bigger total audience, the overlap in audiences of the two, and whether exposure to one meme inhibited the spread of the other.

Gaurav Paruthi

By studying the impact of each message in terms of the number of people who transition between states of interaction and exposure as a result of the message, this research provided raw material for Resnick’s interactive RumorLens tool. 

RumorLens is a project developed by Resnick that analyzes the spread of rumors on Twitter, allowing users to submit a tweet that characterizes a topic that interests them. The application will retrieve additional tweets related to the subject and prompt user feedback to classify results as propagating, debunking or unrelated to the original rumor. It will then use a text classifier to garner more widespread results.

Paruthi’s paper, “Understanding Lending Behaviors on Online Microlending Platforms: The Case for Kiva," analyzed the relationship between the information that lenders can explore in online microlending platforms and the lending activity that information generates. Microlending institutions focus on providing financial services to low-income citizens who typically lack access to traditional banking systems. 

The study examined the Kiva platform, which allows individuals to search through borrower requests for business, school, energy, or other types of loans and lend any amount of money they wish. Paruthi looked at the ratings of microfinance institutions, loan characteristics, and lending terms and found that lenders appear to lend more to highly-rated institutions and that smaller, homogenous teams seem to drive more lending activity and achieve larger lending agreements. 

ICWSM’s overall theme includes research in new perspectives in social theories, as well as computational algorithms for analyzing social media. ICWSM 2014 was held on the U-M campus and co-chaired by Adar and Resnick.

Posted May 21, 2015