An exploration of coronavirus-related online antisemitism in Hungary using quantitative topic model and qualitative discourse analysis

Authors

DOI:

https://doi.org/10.17356/ieejsp.v7i3.801
Abstract Views: 312 PDF Downloads: 133

Keywords:

coronavirus, antisemitism, conspiracy theories, LDA topic modeling, Natural Language Processing (NLP), annotation

Abstract

Soon after the outbreak of the pandemic, antisemitism connected to the coronavirus appeared in the world. In our research we analyzed a large Hungarian online text corpus from December 1, 2019, to July 10, 2020 to examine whether coronavirus-related antisemitism was present in the Hungarian online space, and if so, what its content was. We differentiated between two layers of communication: the professionalized layer represented by online articles, and the lay one represented by comments and posts. After providing the conceptual background regarding conspiracy theories and conspiratorial- and coronavirus-related antisemitism, we present the mixed-method approach that we employed. This approach includes quantitative LDA topic models, human annotation, and the qualitative analysis of various discourses. Our research indicates that coronavirus-related antisemitism appeared in the Hungarian online space at the very beginning of the pandemic. However, at this time, until July, it was present almost solely at the lay level. Its content was mainly related to various tropes (conspiracy theories) about Jews. However, additional content was also identified. Based on our results and international examples, we propose a comprehensive typology that proved to be a suitable means of analyzing coronavirus-related antisemitic content.

Downloads

Published

2021-12-29