A Computational Framework for Multi-Modal Social Action Identification

TL;DR

This paper introduces a computational framework for identifying social actions using multi-modal data, exemplified by an open-source event detection tool analyzing geo-tagged Tweets to understand collective actions globally.

Contribution

It presents a novel framework combining statistical machine learning with social media data for fine-grained social action detection and analysis.

Findings

Successfully built an open-source event detection tool

Analyzed over 600 million geo-tagged Tweets

Enabled detailed understanding of peaceful and violent actions

Abstract

We create a computational framework for understanding social action and demonstrate how this framework can be used to build an open-source event detection tool with scalable statistical machine learning algorithms and a subsampled database of over 600 million geo-tagged Tweets from around the world. These Tweets were collected between April 1st, 2014 and April 30th, 2015, most notably when the Black Lives Matter movement began. We demonstrate how these methods can be used diagnostically-by researchers, government officials and the public-to understand peaceful and violent collective action at very fine-grained levels of time and geography.