Examining the COVID-19 Infodemic on Twitter: A Social Network Analysis in the Context of Thailand

บทคัดย่อ

Previous research on the COVID-19 infodemic has focused on the Western world and a limited time frame. This study aims to bridge the gap by examining the infodemic in a different context - Thailand - over a longer period, from December 31, 2019 to July 31, 2021. The study’s objectives are to: understand how COVID-19 information pollution is spread on Twitter, assess the effectiveness of counter-narratives in reaching users, and identify the most common types of information pollution and trends. Content, sentiment, and social network analyses were conducted to achieve the study's objectives. The results showed that five categories of disinformation were the most common in the dataset: politics (45.70%), medical information (21.31%), vaccine_politics (16.33%), conspiracy_theory (7.68%), and vaccine_medical_info (6.28%). Most nodes interacted with information pollution (59.51%). Only a small proportion of the nodes engaged with debunking/fact-checked messages (16.87%) or both information pollution and debunking/fact-checked messages (23.61%). The results also revealed that the communication network is not completely isolated, as there are nodes that are well-connected to both information pollution and debunking/fact-checked messages. This suggests that users may be exposed to diverse content, even if they are primarily interacting with information pollution. Understanding the problem in its actual context could lead to the development of appropriate and effective responses to the current and future infodemic.