Aggressive Language Identification in Social Media using Deep Learning

Abstract

The increasing influence from users in social media has made that Aggressive content propagates over the internet. In a way to control and tackle this problem, recent advances in Aggressive and offensive language detection have found out that Deep Learning techniques get good performance as well as the novel Bidirectional Encoder Representations from Transformer called BERT. This work presents an overview of Offensive language detection in English and the Aggressive content detection using this novel approach from Transformer for the case study of Mexican Spanish. Our preliminary results show that pre-trained multilingual model BERT also gets good performance compared with the recent approaches in Aggressive detection track at MEX-A3T.