• Medientyp: E-Artikel
  • Titel: The News Cycle's Influence on Social Media Activity
  • Beteiligte: Yates, Andrew; Joselow, Jonah; Goharian, Nazli
  • Erschienen: Association for the Advancement of Artificial Intelligence (AAAI), 2021
  • Erschienen in: Proceedings of the International AAAI Conference on Web and Social Media, 10 (2021) 1, Seite 735-738
  • Sprache: Nicht zu entscheiden
  • DOI: 10.1609/icwsm.v10i1.14761
  • ISSN: 2334-0770; 2162-3449
  • Entstehung:
  • Anmerkungen:
  • Beschreibung: While much work has studied the problem of identifying real-world trends based on social media, none has attempted to explicitly model the news cycle's influence on this social media activity. In this work we attempt to model the news cycle's influence on Twitter activity in the context of "news-centric events." We present a model for estimating the number of tweets posted in response to a news event and propose a method for creating an appropriate ground truth. We find that, although our method is sensitive to variations in the amount of training data, we are able to predict future Twitter activity with reasonable accuracy.