Exploring Twitter streams for Social and Economic impact of COVID in France. – XTCOVIF

Social media have promoted new forms of social interaction, and have changed the way people communicate and interact. People use social media to report the latest news, but also to express their opinions and feelings about real-world events. Users show particular interest in emergency situations such as natural disasters and pandemics. With the worldwide spread of the COVID-19 infection, individual activity on social media platforms has increased. Each day, users post millions of messages related to the pandemic.
In this project, we will study the social and economic impacts of COVID-19 using data obtained from social media. We will focus on 6 different societal issues related to the outbreak of COVID-19, namely (1) people's sentiments and emotions, (2) the decline of tourism, (3) the trust that citizens show in governments, (4) the evolution of language, (5) the increase in racism and xenophobia, and (6) the impact of COVID-19 on population mobility.
The proposed project will provide significant socio-economic contributions. Specifically, the outcomes of the proposed research could be valuable in policy planning and management, allowing the French government to gain a clearer picture of the situation and adopt better measures for the prevention of infection as well as for improving citizens' well-being. Certainly, these tools and results will be useful in the case of a second wave of the pandemic.
?he project will also create valuable resources which will allow researchers and policymakers to study the COVID-19 crisis from a social perspective and to analyze the human behavior and information spreading during the pandemic.
From month 1 to month 3, we plan to start the project by preprocessing the dataset and generating resources. Then from month 3 to month 6, we will use these resources for categorizing tweets based on societal issues. The final and most crucial step is to produce insights into the impact of COVID-19 on society, which we will be focusing on from month 5 to month 12. Our team members’ prior expertise on the involved topics and methods, including but not limited to twitter content classification, twitter event detection, data visualization, representation learning for short texts and graph neural networks for mobility data, will enable us to produce the desired tools, results and resources in a timely manner, befitting the emergency character of the global pandemic.