Toward Predicting Active Participants in Tweet Streams: A case study on Two Civil Rights Events

Online social media arouse much research interest in recent years. In contrast to previous work that focused on the detection of emerging topics, this paper undertakes the prediction of active users in online social events, which is so far rarely explored. This prediction task is formulated as a binary classification problem that built on real-world tweet streams, respectively referring to Ferguson event and New York Chockhold event in this paper. Then, a comprehensive user feature system is designed to characterize the users participated in the events, which includes not only basic statistical characteristics or image-pixel-level features, but also some emotional features and personality features. Next, the Weighted Random Forest (Weighted-RF) classifier is adopted to solve the classification problem. Based on the user feature system and the classifier, the experience of a previous event can be archived and applied to the prediction of later similar events. Experimental results show that the Weighted-RF trained by samples of Ferguson event can effectively predict active users in NYC event, with an AUC value around 0.8392. Besides, the image-content based personality model contributes to the quantitative analysis of population portrait, which brings new perspective to understand the participants of events.