Elucidating the interplay of COVID-19 epidemic and social dynamics via Internet media in Germany
Image and Language Processing, Computer Graphics and Visualisation, Human Computer Interaction, Ubiquitous and Wearable Computing
Empirical Social Research
Epidemiology and Medical Biometry/Statistics
Final Report Abstract
Understanding the social aspects of infectious diseases, highlighted by the COVID-19 pandemic, has become a crucial concern. We were investigating the feasibility of utilizing diverse datasets (Telegram, Google, Twitter, news, etc.) to address various infection-related issues, such as COVID-19 vaccines, conspiracy theories, anti-sanitarian protests, M-pox, long-COVID, among others. This investigation was facilitated through the application of a range of methodologies, including time series analysis, casual models, classic natural language processing and so-called deep learning, spatial models, and social network analysis. The research emphasized the distinction between infoveillance, which primarily focuses on the analysis of web content for predicting biological phenomena, and infodemiology, which involves studying the demand and supply trajectory of information, particularly through search engine queries and social/traditional media content creation or commenting, as demonstrated during the COVID-19 pandemic. The findings emphasized the significance of various data sources in infoveillance and infodemiology, with some sources proving more useful for each approach. The project contributed significantly to the field, collaborating with numerous national and international organizations, and sharing knowledge and expertise with a wide array of partners. Key contributions of the research included enhanced vaccine monitoring measures, early warning of One Health events, and a better understanding of social processes triggered by infection-related issues. Moreover, the research shed light on the insights into social behavior during pandemics, the dynamics of polarization in the context of biology, the spread of misinformation, and the implications for public health policy. This comprehensive understanding could play a pivotal role in formulating effective communication strategies and fostering cohesive societal responses during times of crisis. The project’s key recommendations involve: • Establishing a national Epidemic Intelligence System in Germany, incorporating non-traditional infoveillance elements (similar to the efforts of Robert Koch Institute, Berlin). • Enhancing the application of modeling and data analytics, particularly AI, in global social media monitoring for improved management of infodemics.
Publications
-
Mild Adverse Events of Sputnik V Vaccine in Russia: Social Media Content Analysis of Telegram via Deep Learning. Journal of Medical Internet Research, 23(11), e30529.
Jarynowski, Andrzej; Semenov, Alexander; Kamiński, Mikołaj & Belik, Vitaly
-
Driving Factors of Polarization on Twitter During Protests Against COVID-19 Mitigation Measures in Vienna. Lecture Notes in Computer Science, 15-26. Springer Nature Switzerland.
Röckl, Marcus; Paul, Maximilian; Jarynowski, Andrzej; Semenov, Alexander & Belik, Vitaly
