Local Activity in Online Communities During the Time of Coronavirus

Social networking sites (SNS) not only give people an opportunity to communicate over large distance and in any circumstances but also to create online communities. In such groups, people discuss a variety of topics, ranging from suggestions about the territory improvements to sharing food recipes. The current pandemic of COVID-19 may affect people's behaviour in different ways. This work analyses apartments and city area communities from social networking site VKontakte, the largest SNS in Russia, using computational text analysis (structural topic modeling). Overall, topics were similar to those identified in previous literature. However, after the introduction of lockdown in Saint Petersburg, some of them were less likely to be observed (e.g. construction), while topics related to coronavirus, protection (masks) and home activities were more likely to be discussed.