The study explains how SemEval 2019 Task 5—which involves identifying hate speech against women and immigrants in Spanish and English comments taken from Twitter—was organized. The job is divided into two related classification subtasks: a primary binary subtask for recognizing hate speech and a more detailed one focused on identifying other characteristics in hateful material, such the aggressive attitude and the harassed target, to determine if the incitement is directed at an individual or a group. With 108 submissions for Subtask A and 70 submissions for Subtask B, from 74 different teams, HatEval has been one of the most popular tasks in SemEval-2019.

https://aclanthology.org/S19-2007

By author

Leave a Reply

Your email address will not be published. Required fields are marked *