Automatic Identification of COVID-19-Related Conspiracy Narratives in German Telegram Channels and Chats

Heinrich, Philipp;Blombach, Andreas;Doan Dang, Bao;Zilio, Leonardo;Schäfer, Fabian;et.al.
(2024) Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation — ISBN: [979-10-95546-34-4], 1932-1943, published

Files

2024.lrec-main.173.pdf
  • Open Access
  • Adobe PDF
  • 629.22 KB
Download

Details

Authors
  • Heinrich, Philipp
    Author
  • Blombach, Andreas
    Author
  • Doan Dang, Bao
    Author
  • Zilio, Leonardo
    Author
  • Schäfer, Fabian
    Author
Show more
Abstract
We are concerned with mapping the discursive landscape of conspiracy narratives surrounding the COVID-19 pandemic. In the present study, we analyse a corpus of more than 1,000 German Telegram posts manually tagged with 14 conspiracy and conspiracy-related narrative labels by three independent annotators. Since emerging narratives on social media are short-lived and notoriously hard to track, we experiment with different state-of-the-art approaches to few-shot and zero-shot text classification. We report performance in terms of ROC-AUC and in terms of optimal F1, and compare fine-tuned methods with off-the-shelf approaches and human performance.
Affiliations

Citations

  • APA
  • Chicago
  • FWB

Heinrich, P., Blombach, A., Doan Dang, B., Zilio, L., Havenstein, L., Dykes, N., Evert, S., & Schäfer, F. (2024). Automatic Identification of COVID-19-Related Conspiracy Narratives in German Telegram Channels and Chats. In Nicoletta Calzolari, Min-Yen Kan, Veronique Hoste, Alessandro Lenci, Sakriani Sakti, Nianwen Xue (ed.), Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (pp. 1932-1943). European Language Resources Association (ELRA) and ICCL. https://doi.org/10.63317/5pdoao8iyoyd