Predicting Influence of User's Twitter Activity

The micro-blogging social platform Twitter is being increasingly used nowadays for real-time sharing of news related to a wide range of events, such as elections, protests, disasters, and other news-intensive incidents. In the recent past, the social activity of twittered with sympathies to different political causes has played a major role in several well-known socio-political events, such as the Arab Spring. In a different context, users' twitter feed has helped in a humanitarian disaster relief and rescue in the wake of the 2012 earthquake in Japan. Vast amounts of real-time as well as historical twitter data can be analysed to monitor targets of interest, identify trends in twitter activity, and predict actions on them. In this paper, we consider a case study related to the on-going large-scale and prolonged anti-government protests marches and demonstrations organized by PTI (Pakistan Tehreek Insaf). We analyses the data on twitter activity to identify key hash-tags and propose a model to identify key influences resulting from user's tweet activity.