Learning to Discover Political Activism in the Twitterverse

Abstract

When analysing social media conversations, in search of the public opinion about an unfolding political event that is being discussed in real-time (e.g., presidential debates, major speeches, etc.), it is important to distinguish between two groups of participants: political activists and the general public. To address this problem, we propose a supervised machine-learning approach, which uses inexpensively acquired labeled data from mono-thematic Twitter accounts to learn a binary classifier for the labels “political activist” and “general public”. While the classifier has a 92 % accuracy on individual tweets, when applied to the last 200 tweets from accounts of a set of 1000 Twitter users, it classifies accounts with a 97 % accuracy. Our work demonstrates that machine learning algorithms can play a critical role in improving the quality of social media analytics and understanding, whose importance is increasing as social media adoption becomes widespread.