JMIR Publications

ABSTRACT

Background:

Coronavirus disease (COVID-19) is a scientifically and medically novel disease that is not fully understood as it needs to be consistently and deeply studied. In the past, research on the COVID-19 outbreak was only able to predict quantity data such as the number of outbreaks, but not infoveillance data.

Objective:

This study aims to understand public perceptions on the trends of the COVID-19 pandemic and uncover meaningful themes of concern posted by Twitter users during the pandemic throughout the world.

Methods:

Data mining on Twitter was conducted to collect a total of 107,990 tweets between December 13 and March 9, 2020. The analysis included time series, sentiment analysis and topic modeling to identify the most common topics in the tweets as well as to categorize clusters and find themes from keyword analysis.

Results:

The results indicate three main aspects of public awareness and concerns regarding the COVID-19 pandemic. Firstly, the study indicated the trend of the spread and symptoms of COVID-19, which was divided into three stages. Secondly, the results of the sentiment analysis and emotional tendency showed that the people had a negative outlook toward COVID-19. Thirdly, topic modeling and themes relating to COVID-19 and the outbreak were divided into three categories, including (1) emergency of COVID-19 impact, (2) the epidemic situation and how to control it, and (3) news and social media reporting on the epidemic.

Conclusions:

Sentiment analysis and topic modeling can produce useful information about the trend of COVID-19 pandemic and alternative perspectives to investigate the COVID-19 crisis which has created considerable public awareness around the world. This finding shows that Twitter is a good communication channel for understanding both public concern and awareness about COVID-19 disease. These findings can help health departments to communicate information as to what the public thinks about the disease.