Surveilling COVID-19 Emotional Contagion on Twitter by Sentiment Analysis

Area of Science:

• Public Health
• Social Media Analysis
• Computational Social Science

Background:

• The COVID-19 pandemic involves significant social media discourse.
• Emotional contagion and the need for new surveillance methods are key concerns.
• Twitter is a major platform for public discussion on global health crises.

Purpose of the Study:

• To analyze the flow and content of COVID-19 related tweets.
• To explore the impact of key COVID-19 events on public sentiment on Twitter.
• To assess the role of social media in understanding public emotional responses during the pandemic.

Main Methods:

• Analysis of 3,308,476 English tweets from January 19 to March 3, 2020.
• Sentiment analysis using a polarity compound score and a transformer-based model.
• Examination of longitudinal trends and user activity related to COVID-19 discussions.

Main Results:

• Negative sentiment in tweets increased over time, particularly after major COVID-19 events.
• Highly active users exhibited increasingly negative sentiment scores.
• Tweet content and sentiment trajectories indicated challenges in emotional management of the global emergency.

Conclusions:

• Integrating social media surveillance into pandemic management can combat emotional contagion.
• Twitter analysis offers a novel approach to disseminate reliable information and build trust.
• Social media platforms are vital for monitoring and supporting public health behaviors and community support.