Three PhD students of our department, Fatima Haouari, Maram Hasanain, and Reem Suwaileh, all supervised by Dr. Tamer Elsayed, have won the “Best Paper Award” on their paper titled “ArCOV19-Rumors: Arabic COVID-19 Twitter Dataset for Misinformation Detection” at the Sixth Arabic Natural Language Processing Workshop (WANLP 2021), the premier venue for Arabic NLP, that was held on April 19th and co-located with EACL 2021 conference.
The paper aims to facilitate research on misinformation detection on social media, during this complex and historical period of our time, by introducing a manually-annotated Arabic dataset that covers tweets spreading COVID-19 related claims. ArCOV19-Rumors supports two levels of misinformation detection over Twitter: verifying free-text claims (called claim-level verification) and verifying claims expressed in tweets (called tweet-level verification). ArCOV19-Rumors is the only Arabic dataset made available to support both claim-level and tweet-level verification tasks in Twitter given the propagation networks of the tweets in general and on COVID-19 in particular. The dataset covers, in addition to health, claims related to other topical categories that were influenced by COVID-19, namely, social, politics, sports, entertainment, and religion.
The paper offers baselines for future research on the problem by presenting benchmarking results using SOTA models of versatile approaches that either exploit content, user profiles features, temporal features, or propagation structure of the conversational threads for tweet verification.
Congratulations to Fatima , Maram , Reem, and Dr. Tamer!