Vicarious Offensive Language Identification
Dr. Marcos Zampieri
KEY INTERESTS
Computational linguistics; Natural language processing
AFFILIATIONS/APPOINTMENTS
Assistant Professor, Department of Information Sciences and Technology, George Mason University
ACADEMIC DEGREES
PhD, Computational Linguistics, Saarland University
VICARIOUS OFFENSIVE LANGUAGE IDENTIFICATION
For over a decade, across hundreds of studies, researchers on offensive language identification asked annotators: “Do you find this social web post offensive?” The project team has recently introduced the notion of vicarious offense, reframing the problem by asking: “Do you think Group A will find this social web post offensive?” This project builds on the notion of vicarious offense applied to content moderation, identifying how vicarious interaction can be leveraged to develop more inclusive, transparent, and trustworthy content moderation processes.