engleski

Natural language processing and statistic: The first six months of the COVID-19 infodemic in Croatia

This research aims to contribute to the answers to the following two questions: how did the infodemic develop in the digital information media, and what was the relationship between the infodemic and the development of the pandemic? We suggest that the first step towards a better understanding of the COVID-19 infodemic is to perform a quantitative analysis of news articles published on online portals. The main goal of this study is to provide an overview of the trends of COVID-19 related articles published on online news portals during the first six months of the pandemic in Croatia. We propose a three-step analysis of texts: (i) analysis of the number of articles related to COVID-19, (ii) investigation of the correlations between the number of articles and the number of daily COVID-19 cases and (iii) linguistic analysis of texts (observing the most frequent keywords and how the trends change during the time). Based on the results of investigating the correlation between the number of published news-articles and the number of daily reported cases of infection shows that the news portals insist on the pandemic as a topic even when the events that generate it are no longer at their peak.

COVID-19 ; infodemic ; crisis communication ; social media ; online portals ; natural language processing

nije evidentirano