How Newspapers Portrayed COVID-19: A Study Based on United Kingdom and Bangladesh

Rizvee, R. A.; Zaber, M.

In this study, we wanted to see how as a representative of a society newspaper portrayed COVID-19. For this purpose, this study considered two countries United Kingdom (UK) and Bangladesh (BD), and analyzed how COVID-19 as an external event was focused in the newspapers. To conduct the analysis, we handpicked a set of covid related feature terms, and using Latent Dirichlet Allocation (LDA) we verified the coherency of the chosen terms. The main finding of this study is that, initially as a new event COVID-19 found huge importance in the newspapers, but with gradual time progression, this became a new normal event and lost its initial insurgence of focus and took a stable condition. We also observed that despite being quite different demographically, geographically, economically along with being affected differently due to covid, UK and BD exhibit quite similar characteristics in portraying covid in newspapers. The decisions were arrived at by applying different types of Natural Language Processing (NLP) tools and statistical analysis. For experimentation, we collected all the published news articles from two newspapers, the Guardian and the Daily Star from January 1, 2020, to March 31, 2021. © 2021, IFIP International Federation for Information Processing.