HateMonitors: Language Agnostic Abuse Detection in Social Media

TL;DR

HateMonitors is a language-agnostic machine learning system that effectively detects hate speech and offensive content in social media, leveraging BERT and LASER embeddings, and achieved top results in a multilingual shared task.

Contribution

The paper introduces HateMonitors, a novel language-agnostic model for abusive content detection using BERT and LASER embeddings, with state-of-the-art performance in a multilingual setting.

Findings

Achieved first place in German sub-task A of HASOC.

Developed a language-agnostic model using BERT and LASER embeddings.

Made the model publicly available for further research.

Abstract

Reducing hateful and offensive content in online social media pose a dual problem for the moderators. On the one hand, rigid censorship on social media cannot be imposed. On the other, the free flow of such content cannot be allowed. Hence, we require efficient abusive language detection system to detect such harmful content in social media. In this paper, we present our machine learning model, HateMonitor, developed for Hate Speech and Offensive Content Identification in Indo-European Languages (HASOC), a shared task at FIRE 2019. We have used a Gradient Boosting model, along with BERT and LASER embeddings, to make the system language agnostic. Our model came at First position for the German sub-task A. We have also made our model public at https://github.com/punyajoy/HateMonitors-HASOC .