DamascusTeam at NLP4IF2021: Fighting the Arabic COVID-19 Infodemic on Twitter Using AraBERT

摘要

In the modern era of computing, the news ecosystem has transformed from old traditional print media to social media outlets. Social media platforms allow us to consume news much faster, with less restricted editing results in the spread of infodemic misinformation at an incredible pace and scale. Consequently, the research on the infodemic of the post's misinformation is becoming more important than ever before. In this paper, we present our approach using AraBERT (Trans former-based Model for Arabic Language Understanding) to predict 7 binary properties of an Arabic tweet about COVID-19. To train our classification models, we use the dataset provided by NLP4IF 2021. We ranked 5th in the Fighting the COVID-19 Infodemic task results with an F1 of 0.664.