Using Deep Learning to Identify Collective Action Events with Text and Image Data from Social Media

• In Event: 4464 - Regular Session on Methodology: Progress and New Approaches

Abstract

Protest event analysis is an important method for the study of collective action and social movements, which typically draws on traditional media reports as the data source. We introduce Collective Action from Social Media (CASM)---a system that uses convolutional neural networks on image data and recurrent neural networks with long short-term memory on text data in a two-stage classifier to identify collective action events occurring offline. We implement CASM on Chinese social media data and identify 142,427 collective action events from 2010 to 2017 (CASM-China). We extensively evaluate the performance of CASM through cross-validation, out-of-sample validation, and comparisons with other Chinese protest datasets. We assess the impact of online censorship, and find that it does not substantially limit our identification of events. Compared to other datasets of protests, CASM-China identifies relatively more rural, land-related protests, but identifies few collective action events related to ethnic and religious conflict.

Authors

• Han Zhang, Princeton University

• Jennifer Pan, Stanford University