Private Analysis of Social Networks

• Activity Number: 627
• Type: Invited
• Date/Time: Thursday, August 7, 2014 : 10:30 AM to 12:20 PM
• Sponsor: Statistical and Applied Mathematical Sciences Institute
• Author(s): Aleksandra Slavkovic*+ and Vishesh Karwa
• Companies: Penn State and Penn State
• Keywords: confidentiality ; differential privacy ; degree sequence ; socia networks ; synthetic graphs

Abstract

Social networks are a prominent source of data for researchers in epidemiology, sociology and many other disciplines and have sparked a flurry of research in statistical methodology for network analysis. While the social benefits of analyzing these data are significant, they often contain sensitive information and their release can be devastating to the privacy of individuals and organizations. In this talk, we give a brief overview of challenges associated with protecting such data, and the problem of releasing summary statistics of graphs needed to build statistical models for networks while preserving privacy of individual relations. We present an algorithm designed to provide utility for statistical inference in random graph models whose sufficient statistics are functions of degree sequences which are released under the framework of differential privacy. Specifically, we focus on the tasks of existence of maximum likelihood estimates, parameter estimation and goodness-of-fit testing for the beta model. The algorithm's performance is evaluated on both the simulated and the real-life datasets. Our algorithm can also be used to release synthetic graphs under the beta model.