COVID-19 Public Sentiment Insights and Machine Learning for Tweets Classification

Abstract

Along with the Coronavirus pandemic, another crisis has manifested itself in the form of mass fear and panic phenomena, fueled by incomplete and often inaccurate information. There is therefore a tremendous need to address and better understand COVID-19's informational crisis and gauge public sentiment, so that appropriate messaging and policy decisions can be implemented. In this research article, we identify public sentiment associated with the pandemic using Coronavirus specific Tweets and R statistical software, along with its sentiment analysis packages. We demonstrate insights into the progress of fear-sentiment over time as COVID-19 approached peak levels in the United States, using descriptive textual analytics supported by necessary textual data visualizations. Furthermore, we provide a methodological overview of two essential machine learning (ML) classification methods, in the context of textual analytics, and compare their effectiveness in classifying Coronavirus Tweets of varying lengths. We observe a strong classification accuracy of 91\% for short Tweets, with the Na\"ive Bayes method. We also observe that the logistic regression classification method provides a reasonable accuracy of 74\% with shorter Tweets, and both methods showed relatively weaker performance for longer Tweets. This research provides insights into Coronavirus fear sentiment progression, and outlines associated methods, implications, limitations and opportunities.

Twitter and Census Data Analytics to Explore Socioeconomic Factors for Post-COVID-19 Reopening Sentiment

“…The present study utilizes public sentiment derived from social media posts as a key variable, and this is supported by extant research which has used sentiment analysis for diverse research purposes such as decision support, education, politics, opinion mining, data visualization, healthcare and hate crimes, and the importance of education, gender sensitivity and motivation [7,22,24,[26][27][28][29][30][31][32]. These studies have used a wide range of methods, tools and languages such as Python and R, and their associated libraries, to estimate sentiment from social media posts.…”

“…COVID-19 was first identified in Wuhan, China, by the end of December 2019, already affected over 10 million people in 213 countries of the world with a fatality rate, that had reportedly almost reached 10% among the closed cases [6]. It is a highly infectious and deadly disease, with widespread transmission and significantly negative impacts on physical, emotional and mental health, economy, and peoples way of living [7,8]. To control the spread of COVID-19, federal government declared statewide emergency and states governments implemented stay-athome-order, imposed restriction on mass gathering and non-essential movements.…”

“…Past research investigated fear and trust sentiments of the people towards reopening the economy in the US using exploratory textual analytics, textual data visualization, and hypothesis testing techniques [8]. Similarly, past research has also investigated and classified sentiments of the users (e.g., positive, negative) towards an item, situation, and system [7]. However, they rarely discussed the underlying socioeconomic factor associations for such sentiments.…”

Twitter and Census Data Analytics to Explore Socioeconomic Factors for Post-COVID-19 Reopening Sentiment

“…COVID-19 was first identified in Wuhan, China, by the end of December 2019, already affected over 10 million people in 213 countries of the world with a fatality rate, that had reportedly almost reached 10% among the closed cases [6]. It is a highly infectious and deadly disease, with widespread transmission and significantly negative impacts on physical, emotional and mental health, economy, and people's way of living [7,8]. To control the spread of COVID-19, federal government declared statewide emergency and states governments implemented stay-athome-order, imposed restriction on mass gathering and non-essential movements.…”

“…has also investigated and classified sentiments of the users (e.g., positive, negative) towards an item, situation, and system [7]. However, they rarely discussed the underlying socioeconomic factor associations for such sentiments.…”