KUISAIL at SemEval-2020 Task 12: BERT-CNN for Offensive Speech Identification in Social Media

TL;DR

This paper presents a BERT-CNN approach for multilingual offensive speech detection, demonstrating improved performance over BERT alone and sharing ArabicBERT models for Arabic language processing.

Contribution

The paper introduces a combined BERT-CNN model for offensive speech identification and shares ArabicBERT, a set of pre-trained Arabic language models, advancing multilingual offensive language detection.

Findings

BERT-CNN outperforms BERT alone in accuracy.

The system achieved top ranks in SemEval-2020 tasks.

ArabicBERT models are shared for community use.

Abstract

In this paper, we describe our approach to utilize pre-trained BERT models with Convolutional Neural Networks for sub-task A of the Multilingual Offensive Language Identification shared task (OffensEval 2020), which is a part of the SemEval 2020. We show that combining CNN with BERT is better than using BERT on its own, and we emphasize the importance of utilizing pre-trained language models for downstream tasks. Our system, ranked 4th with macro averaged F1-Score of 0.897 in Arabic, 4th with score of 0.843 in Greek, and 3rd with score of 0.814 in Turkish. Additionally, we present ArabicBERT, a set of pre-trained transformer language models for Arabic that we share with the community.