An empirical study on Twitter's use and crisis retweeting dynamics amid Covid-19

This study conducts an analysis on topics of the most diffused tweets and retweeting dynamics of crisis information amid Covid-19 to provide insights into how Twitter is used by the public and how crisis information is diffused on Twitter amid this pandemic. Results show that Twitter is first and foremost used as a news seeking and sharing platform with more than $$ of the most diffused tweets being related to news and comments on crisis updates. As for the retweeting dynamics, our results show an almost immediate response from Twitter users, with some first retweets occurring as quickly as within 2 s and the vast majority $$ of them done within 10 min. Nearly $$ of the retweeting processes could have $$ of their retweets finished within 24 h, indicating a 1-day information value of tweets. Distribution of retweeting behaviors could be modeled by Power law, Weibull, and Log normal in this study, but still there are $$ original tweets whose retweeting distributions left unexplained. Results of retweeting community analysis show that following retweeters contribute to nearly $$ of the retweets. In addition, the retweeting contribution of verified Twitter users is significantly $$ different from that of unverified users. A similar significant $$ difference is also found in their rates of verified retweeters, and it has been shown that verified Twitter users enjoy seven times as high value as that of unverified users. In other words, users with the same verification status are more likely to get together to diffuse crisis information.