Covid-19 vaccines in Italian public opinion: Identifying key issues using Twitter and Natural Language Processing

Area of Science:

• Public Health
• Social Media Analysis
• Computational Social Science

Background:

• The COVID-19 pandemic necessitated a rapid vaccination campaign across the European Union starting December 2020.
• Social media platforms serve as crucial channels for understanding public perception of health initiatives.
• Monitoring public sentiment towards COVID-19 vaccines is vital for campaign success.

Purpose of the Study:

• To analyze public sentiment and identify key topics related to COVID-19 vaccination in Italy using Twitter data.
• To track the progression of public opinion during the initial months of the vaccination campaign.
• To provide insights for policymakers to enhance vaccination promotion strategies.

Main Methods:

• Collected over 71,000 tweets from Italian users mentioning COVID-19 vaccines (February-May 2021).
• Employed VADER for spatial and temporal sentiment analysis to detect sentiment shifts.
• Utilized Dynamic Latent Dirichlet Allocation (DLDA) to identify dominant discussion topics.

Main Results:

• Sentiment analysis showed significant fluctuations, often linked to news about vaccine side effects.
• The Oxford-AstraZeneca vaccine received the least positive reception among different vaccine brands.
• Key topics included vaccination plan information, perceived usefulness, and concerns regarding vaccine risks, side effects, and safety.

Conclusions:

• This study offers a novel analysis of Italian Twitter users' opinions on COVID-19 vaccination.
• Public sentiment is dynamic and influenced by reported vaccine side effects and specific vaccine brands.
• Findings can inform public health strategies for effective vaccine communication and promotion.