Computer Vision

#134: ISIS Radicalization, Counter-Extremism, and Visual Propaganda on Social Media, with Dr. Tamar Mitts

October 24, 2021 Academics

Dr. Tamar Mitts, Assistant Professor of International and Public Affairs at Columbia University, shares her research on ISIS radicalization and it’s relationship to social isolation, anti-Muslim sentiment, and counter-extremism programs. We discuss Dr. Mitts’ Twitter dataset that uses spatial algorithms to identify ISIS sympathizers’ locations and machine learning to identify pro-ISIS sentiment. Toward the end of the episode, we discuss how computer vision tools such as Amazon’s Rekognition API can be used to detect violent imagery in ISIS propaganda. 


Here are the studies we discuss in the episode: 


From Isolation to Radicalization: Anti-Muslim Hostility and Support for ISIS in the West  (2019)

Countering Violent Extremism and Radical Rhetoric (2021)

Studying the Impact of ISIS Propaganda Campaigns (Forthcoming)

#133: Covid Mask Wearing in Politicians’ Social Media Images and Emotions in German Politics, with Dr. Mirya Holman

October 10, 2021 Academics

Dr. Mirya Holman, Associate Professor of Political Science at Tulane University, shares her latest research on how politicians depict mask wearing through their social media images. We discuss how computer vision can be used to detect masks in images, as well as what factors correlate with politicians’ depicting masks. Later in the episode, we discuss another recent study by Dr. Holman, where emotions in the facial expressions and vocal pitch of German politicians were analyzed during election debates. 

 

Here’s a link to that study: 

Gender, Candidate Emotional Expression, and Voter Reactions during Televised Debates (2021)

 

And here’s a link to Mirya Holman’s Aggressive Winning Scholars (#MHAWS) Newsletter!