Multi-Task Learning for Detection of Aggressive and Violent Incidents from Social Media

Abstract

In this paper, we participate in the task of Detection of Aggressive and Violent INCIdents from Social Media in Spanish (DA-VINCIS). We apply a multi-task learning network, MT-DNN to train users’ tweets on their text embeddings from pre-trained transformer models. In the first subtask, we obtained the best F1 of 74.80%, Precision of 75.52%, and Recall of 74.09%. Meanwhile, F1 of 39.20%, Precision of 37.79%, and Recall of 43.88% are results in the second subtask.