A New User Recommendation Model Within the Context of the Covid-19 Pandemic

Event-based social networks provide people with fantastic platforms to improve their relationships and make friends through offline and online activities. Predicting the event attendance of users is a challenging problem and solved by many techniques. Recently, the outbreak of Covid-19 changes the ways that users participate in events, from offline to online. In this paper, we study the problem of user recommendation within the context of the Covid-19 pandemic. To address this problem, we first analyze the information of events to obtain three factors, i.e., content, time, and location. Then, we propose a new recommendation model to compute scores of new events with respect to participated events of each user. Finally, the top N events with the highest scores are recommended to the user. Extensive experiments were conducted on a real Meetup event dataset, and the results have shown that our model outperforms comparison methods.