Big Data, Big Issues: Using Public Opinion Theory, Machine Learning, and Social Network Analysis to Explore Opinions and Information Flow Across Traditional and Social Media
Amy Becker (Mass Communication) & Sidd Kaza (Computer and Information Science)
Social media applications on the web are seeing explosive growth due to the increased access to computing resources. According to an August 2011 study by the Pew Research Center’s Internet & American Life Project, 65% of adults use social networking sites and 43% of adults visit these sites on a daily. This is the single largest shift of users to an application (over half a billion users over all online social networks) since the move to the world-wide-web.
The rapidly expanding social media environment has piqued the interest of scholars in a wide range of disciplines yet current research efforts exist in independent silos. Social science researchers focus on the theoretical underpinnings, ethical implications, and consequences of this shift toward a more interactive emerging technology, while computer scientists focus on refining the algorithms that highlight patterns in the data and make sense of this vast array of social media content. In this interdisciplinary project, we hope to pair current thinking in the social sciences on the impact of social media on civic engagement with recent research in computer science on social network analysis (SNA), machine learning, and natural language processing. Specifically, our research questions are as follows:
How can we track the flow of information in traditional and social media outlets while focusing on specific issues of public interest?
How can we identify the influential individuals who are opinion leaders in social media?
Is there a difference in the way interest groups focused on different sides of issues use social media?
Impact on Students
Our graduate student, Ranjan Vaidyanathakumar, is completing his M.S. thesis in Computer Science on Sentiment Analysis of Micro-blog Data as part of this project. An undergraduate student, Andrea Hackl is working with the team on newspaper article coding. Lessons learned in this research project will also be applied in the planned revamping of the research methods course MCOM 490.
The paper "Big Data, Big Issues: Applying Public Opinion Theory, Machine Learning, and Large-Scale Text Analysis to Explore Issue Opinions and Information Flow Across Traditional and Social Media," written by Amy Bree Becker, Siddharth Kaza, and Andrew B. Goldberg has been accepted for presentation for the 2013 APSA Pre-Conference on Political Communication on Wednesday, August 28th, 2013 at the University of Illinois at Chicago.
Amy B. Becker, (2012). Determinants of public support for same-sex marriage: Generational cohorts, social contact, and shifting attitudes. International Journal of Public Opinion Research. DOI: 10.1093/ijpor/EDS002
Xuan Liu, Siddarth Kaza, Pengzhu Zhang, and Hsinchun Chen (2011). Effect of Inventor Status on Innovation Diffusion within Institutions: A Study on SCI Literature from China, Russia, and India, Journal of the American Society for Information Science and Technology (JASIST), 62 (6), pp 1166-1176, DOI: 10.1002/asi.21528
Siddarth Kaza and Hsinchun Chen (2010). Identifying High-status Nodes in Knowledge Networks, Annals of Information Systems, 12, pp. 91-108. DOI: 10.1007/978-1-4419-6287-4_6
Amy B. Becker, A.B., & Scheufele, D.A. (2011). New voters, new outlook? Predispositions, social networks, and the changing politics of gay civil rights. Social Science Quarterly, 92(2), 324-345. DOI: 10.1111/j.1540-6237.2011.00771.x.