Media Coverage and Public Perception of Distance Learning During the COVID-19 Pandemic: A Topic Modeling Approach Based on BERTopic

The current study proposes a topic modeling approach for analyzing press coverage and public perception of distance learning during the COVID-19 pandemic. We evaluate the applicability of a novel approach for neural topic modeling based on transformer-based language models. Our methodology is tested empirically on a large sample of news and discussions in various social and news media platforms to derive valuable insights on press coverage and public perception of COVID-19 impact on the education system in Bulgaria. The study outlines key advantages of using BERTopic in analyzing big data. Our work contributes to the body of literature devoted on value creation through big data and text analytics utilization in the public sector.