A crucial point about the COVID-19 pandemic is not only its impact on the health of population, but moreover the flow of information related to it. In the recent months lots of false information about the disease, measurements taken by the government and backgrounds regarding the pandemic was spread and contributed to a partially aggressive climate in discussions about the topic. As a case study, we will examine the structure, language and contents, as well as the user networks of respective debates on Twitter. The goal is to find clusters in the network that behave similar, identify patterns of usage and language relating to specific beliefs and give general insights into the behaviour of distinct groups.

To successfully start and conduct the research project, further clarifications with the supervisor Raphael Wimmer and the second group focusing on the Twitter topic have to be made. For that purpose, a team meeting will be planned, where differention between both groups as well as the concrete focus for our group will be discussed.

* Kay Brinkmann | kay.brinkmann@stud.uni-regensburg.de | 3rd semester, M.Sc. Media Informatics * Luis Moßburger | luis1.mossburger@stud.uni-regensburg.de | 2nd semester, M.Sc. Media Informatics * Felix Wende | felix.wende@stud.uni-regensburg.de | 3rd semester, M.Sc. Media Informatics

This topic fits our research interest perfectly. In „Digital Humanities“ we compared two online forums for depression via Natural Language Processing (NLP) methods, like Sentiment Analysis or Named Entity Recognition. In the „Praxisseminar“, we did User Research for the „Mittelbayerische Zeitung“, part of which was a Network Analysis. In this project, we want to combine those methods and make sense of COVID-19-related debates on Twitter.