Detecting Emotional Evolution on Twitter during the COVID-19 Pandemic Using Text Analysis

Area of Science:

• Social Sciences
• Computational Social Science
• Public Health

Background:

• The COVID-19 pandemic, caused by SARS-CoV-2, presented an unprecedented global crisis.
• Governments implemented widespread restrictions, impacting daily life and economic activities.
• Social media platforms like Twitter became crucial for public expression of emotions and opinions.

Purpose of the Study:

• To analyze the emotional impact of the COVID-19 pandemic on Twitter users in Spanish-speaking countries.
• To track the evolution of public sentiment regarding the pandemic over 13 months (March 2020 - March 2021).
• To understand how pandemic-related issues influenced population emotions and opinions.

Main Methods:

• Utilized a dataset of 3 million tweets from Chile, Mexico, Peru, and Spain.
• Employed the EmoWeb framework to analyze the dynamic sentiment of pandemic-related keywords.
• Conducted a temporal analysis of emotional variations in public discourse.

Main Results:

• The study identified significant shifts in public emotions on Twitter throughout the pandemic.
• Results demonstrate varied impacts of the pandemic and its consequences across different countries.
• The evolution of opinions on various COVID-19 topics was mapped over the study period.

Conclusions:

• The COVID-19 pandemic profoundly affected the emotional landscape of social media users in Spanish-speaking regions.
• Twitter sentiment analysis provides valuable insights into public perception and emotional responses to global health crises.
• Understanding these emotional dynamics is crucial for public health communication and policy-making.