Using Natural Language Processing to Automate Detection of Targeted Attacks in Political Tweets

• In Event: Standing ePanel: Political Rhetoric

Brief Overview

We develop a classification model to detect attacks in U.S. House members’ Tweets based on the text and contextual data such as political party. We apply this model to a novel dataset, assessing whether 1.25 million Tweets contain political attacks.

Authors

• Nicholas I Deas, Clemson University

• Jacob Sargent, Clemson University

• Spencer A Shellnutt, Clemson University