Identifying Influential Twitter Users in the 2011 Egyptian Revolution

Recent international events surrounding contentious political environments have uncovered a new utility for social media. Communities now use resources such as Facebook and Twitter to quickly spread information and project influence amongst potentially geographically disparate people. In this work, we investigate Twitter activity in Egypt during the 2011 protests and revolution, and introduce a model to automatically ascertain key individuals within these networks. The model takes advantage of a more sparse network on Twitter than the traditional follower/following network by leveraging direct communications. Furthermore, we employ a measure of alpha centrality, which incorporates both directionality of network connections and a measure of external importance. The model is applied to topic-based communities within Twitter rather than previously introduced measures of influence that focus on the cascading spread of single messages or broad, topicinvariant measures. Results indicate a model successful at automatically identifying users that are active and influential within a given community, agreeing well with heuristics and comparable to other influence models but with particular advantages such as tunability and robustness