Paper short abstract:

The current work focuses on the application of machine learning and natural language processing to discovering the underlying narrative frameworks in social media discussions that focus on the Covid-19 pandemic, its causes and its spread.

Paper long abstract:

In early 2020, the world realized that the respiratory illness which had first been identified in the far east at the end of 2019 was not only highly contagious and dangerous, but also had made the fairly easy jump to the global stage. Since so little was known about the virus, and since various governments had either directly or indirectly caused people to have little or no trust in readily accessible information sources, the environment was well primed for the emergence of numerous unofficial narratives intending to explain the pandemic. With social media being broadly available, and with its global reach presenting the possibility of a near-instantaneous communication channel, the spread of stories and story parts along these channels was rapid. Consequently, social media became a heavily used medium for the circulation and negotiation of beliefs concerning the virus. While in many past crises the problem for understanding has been access to too little information, the Covid-19 pandemic may actually have swung in the opposite direction, with too much hard to evaluate information being available. In our work, we attempt to discover and track the emergence of narrative frameworks--a dynamic network graph representing the actants and their interactions--across multiple social media sites (Reddit, 4Chan, 8kun, Facebook). The goal is to understand the how the narrative frameworks emerge, stabilize and solidify, as well as to see if we can identify structural features of these networks that differentiate a narrative complex such as a conspiracy theory from other types of narrative frameworks.