Skip to main content
Bild: Sara Kurfess (unsplash)

How do tweets about COVID-19 spread? The COVID-19 Retweet Prediction Challenge should provide possible answers. Participants in the competition are asked to model the retweet behavior on Twitter. TweetsCOV19, a semantically annotated corpus of tweets about the COVID-19 pandemic, is intended to help with this. It is extracted from the long-term Twitter archive TweetsKB and aims to capture the online discourse on various aspects of the pandemic and its social impact. Metadata information about the tweets as well as extracted entities, moods, hashtags and user mentions will be disclosed using established vocabularies and W3C standards. The dataset is continuously updated and currently consists of more than 8 million tweets posted by more than 3 million users and reflects the social discourse on COVID-19 on Twitter from October 2019 to April 2020.

 The competition takes place between July 1 and August 31, 2020 as part of the CIKM2020 AnalytiCup. The finalists will be invited to present their solutions during the CIKM 2020 AnalytiCup workshop in October 2020, which will be held together with the CIKM conference. The winners of the COVID-19 Retweet Prediction Challenge will receive prizes worth €2,500, which will be made available by the L3S Research Centre.