Premium
A preliminary investigation of COVID ‐19 transmission in the United States by incorporating social media sentiments
Author(s) -
Tan Tianyi,
Huang Teng,
Wang Xi,
Zuo Zhiya
Publication year - 2020
Publication title -
proceedings of the association for information science and technology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.193
H-Index - 14
ISSN - 2373-9231
DOI - 10.1002/pra2.370
Subject(s) - covid-19 , social media , transmission (telecommunications) , government (linguistics) , set (abstract data type) , pandemic , state (computer science) , political science , internet privacy , computer science , telecommunications , world wide web , medicine , virology , disease , philosophy , pathology , algorithm , outbreak , infectious disease (medical specialty) , programming language , linguistics
COVID‐19 has now become a global pandemic. During the widespread of COVID‐19, Twitter, as an online social media platform, has been a preferred channel for interaction and communication. As a result, it provides huge amount of information from which latent signals such as sentiments can be mined for a better understanding of COVID‐19 transmission patterns. As a preliminary attempt, we reveal a strongly positive zero‐order correlation between sentiments of tweets and COVID‐19 confirmed cases in U.S. Considering the unique hierarchical structure of the U.S. government, state governments exert their own power to issue public health policies. Indeed, there are different patterns of correlations between sentiments and COVID‐19 confirmed cases, affirming that country‐level characteristics suppress that of state‐level. Diving deeper into the textual content of COVID‐19 related tweets, there manifests a diverse set of topics which in turn lead to dispersed sentiments. Our preliminary investigation paves the way for a finer‐grained analysis of the COVID‐19 transmission and social media activities by considering varying situations across states and topics.