Ongoing Research

Disinformation in Social Media and Violent Political Behavior

This project utilizes social media data including text, images, and other digital trace data, with various NLP and machine learning methods, to develop psychometric profiles of clusters of anonymized social media users which will be used to understand which users tend to spread disinformation in social media networks and for what reasons. Then, using these profiles, we will attempt to determine if psychological factors that predict the spread of disinformation also correlate with violent offline political behavior, such as engaging in protest activity.

A Benchmark Dataset for Detecting and Explaining Implicit Hate Speech

Hate speech has grown significantly on social media, causing serious consequences for victims of all demographics. Despite much attention being paid to characterize and detect discriminatory speech, most work has focused on explicit or overt hate speech, failing to address the pervasive implicit hate speech that uses coded or indirect language to disparage a protected group or individual. To fill this gap, this work introduces a theoretical taxonomy of implicit hate speech and a benchmark corpus with fine-grained labels for each message and its implication. We present systematic analyses of our dataset and results achieved by state-of-the-art baselines in detecting and explaining implicit hate speech, as well as discuss certain challenges that existing approaches struggle with, suggesting that this dataset can serve as a useful research benchmark for understanding implicit hate speech online.