Decoupled Data for Privacy Preserving Record Linkage with Error Management

Presenter: Hye-Chung Kum, School of Social Work and Department of Computer Science, UNC-Chapel Hill

Abstract:
Computational social science is an emerging research area at the intersection of computer science, social sciences, and statistics, in which quantitative methods and computational tools are used to answer complex questions about people and our society. Data-driven studies of our society require a well-integrated data infrastructure of data collected at various times in our lives. From the day we are born until our death, almost all of our activities leave traces in various government databases. Indeed, these government information systems continuously generate data about all aspects of our society, much like the satellites we use to monitor our physical surroundings. If we can build a well-integrated data system that could federate much of these databases on an ongoing basis, the data infrastructure could hold the collective representation of our society, our social genome. Designed, built, and properly managed by experts in IT, data, privacy, security, policy, and content, under proper protocols and oversight, such a social genome data infrastructure could transform social sciences while still protecting individual privacy, just as well-maintained satellite data has transformed astronomy. The transformative benefits to social science are obvious; as are the potential violations of privacy from a poorly designed system. In this talk, we will present

• a working definition of computational social science
• the hierarchy for computational social science that demonstrates how data, method, information, knowledge, and action relate as well as how different types of research in computational social science relate to each other
• an architecture for a secure social genome data infrastructure that can support computational social science
• and initial findings in privacy preserving record linkage systems required for privacy protection in such an infrastructure

Speaker Bio

Hye-Chung Kum is a research assistant professor in the School of Social Work with a joint appointment in the department of computer science at the University of North Carolina at Chapel Hill (UNC-CH). She received her M.S. (1997) and Ph.D. (2004) in computer science from UNC-CH. While completing her doctoral degree, she also completed her M.S.W. (1998) from UNC-CH. She has done extensive research on sequential pattern mining, data mining, digital government, computational social science, and the use of KDD technology on administrative data from social welfare for program evaluation and policy analysis.