AuthorsD. T. Schroeder
EditorsJ. Langguth
TitleExplaining News Spreading Phenomena in Social Networks
AfilliationCommunication Systems
Project(s)Enabling Graph Neural Networks at Exascale, UMOD: Understanding and Monitoring Digital Wildfires
StatusPublished
Publication TypeTalk, keynote
Year of Publication2021
Location of TalkHändlerlogo BI Norwegian Business School
Abstract

Digital wildfires are fast spreading online misinformation phenomena with the potential to cause harm in the physical world. They have been identified as a considerable risk to developed societies which raised the need to better understand online misinformation phenomena to mitigate that risk. We approach the problem from an interdisciplinary angle with the aim of using large scale analysis of social network data to test hypotheses about the behavior of social network users interacting with misinformation. We discuss state of the art techniques for capturing large volumes of communication data from social networks such as Twitter as well as collections of news such as GDELT. Based on that we describe new methods on how the reach as well as the typical target audience of media and social network participants can be measured. Doing so allows the testing of hypotheses such as the existence of filter bubbles through the use of large amounts of real-world data. Finally we discuss how the detection of anomalies in the typical news spreading patterns can be used to detect disinformation campaigns and digital wildfires.

 
Citation Key28043