Finding Clusters of Similar-minded People on Twitter Regarding the Covid-19 Pandemic

TL;DR

This paper introduces two clustering methods to identify groups of Twitter users with similar opinions on Covid-19 in Germany, aiming to understand public debate and combat misinformation.

Contribution

It presents a novel network-based and hashtag-based clustering approach for analyzing opinion groups on Twitter regarding Covid-19.

Findings

Clusters correspond to different public opinion groups.

The two methods produce different clusters due to filtering differences.

Both methods successfully identify opinion-based user groups.

Abstract

Two clustering methods to determine users with similar opinions on the Covid-19 pandemic and the related public debate in Germany will be presented in this paper. We believe, they can help gaining an overview over similar-minded groups and could support the prevention of fake-news distribution. The first method uses a new approach to create a network based on retweet relationships between users and the most retweeted users, the so-called influencers. The second method extracts hashtags from users posts to create a "user feature vector" which is then clustered, using a consensus matrix based on previous work, to identify groups using the same language. With both approaches it was possible to identify clusters that seem to fit groups of different public opinions in Germany. However, we also found that clusters from one approach can not be associated with clusters from the other due to filtering steps in the two methods.