Social-media-based public policy informatics: Sentiment and network analyses of U.S. Immigration and border security

Social media provide opportunities for policy makers to gauge pubic opinion. However, the large volumes and variety of expressions on social media have challenged traditional policy analysis and public sentiment assessment. In this article, we describe a framework for social-media-based public policy informatics and a system called “iMood” that addresses the needs for sentiment and network analyses of U.S. immigration and border security. iMood collects related messages on Twitter, extracts user sentiment and emotion, and constructs networks of the Twitter users, helping policy makers to identify opinion leaders, influential users, and community activists. We evaluated the sentiment, emotion, and network characteristics found in 909,035 tweets posted by over 300,000 users during three phases between May and November 2013. Statistical analyses reveal significant differences in emotion and sentiment among the 3 phases. The Twitter networks of the 3 phases also had significantly different relationship counts, network densities, and total influence scores from those of other phases. This research should contribute to developing a new framework and a new system for social-media-based public policy informatics, providing new empirical findings and data sets of sentiment and network analyses of U.S. immigration and border security, and demonstrating a general applicability to different domains.