An Impossible Dialogue! Nominal Utterances and Populist Rhetoric in an Italian Twitter Corpus of Hate Speech against Immigrants

Gloria Comandini, Viviana Patti
2019 Proceedings of the Third Workshop on Abusive Language Online   unpublished
human-in-the-loop solutions that rely on crowd workers' ratings along with automated moderation, and embedding the evaluations of models into the cultural fabric. Thus, human ratings in the context of toxicity in language raise important questions around the various socio-cultural biases that affect those ratings, but also on the impact it has on the psychological safety of the raters themselves. In order to situate our conversation around this theme, we have confirmed four keynote speakers and
more » ... panelists who are leading experts on content moderation, crowd work, and the impact of algorithmic solutions on people:
doi:10.18653/v1/w19-3518 fatcat:liwo47b4kzblhifxy7gn4lrifq