Social media mining under the COVID-19 context: Progress, challenges, and opportunities

Area of Science:

• Public Health
• Data Science
• Computational Social Science

Background:

• Social media platforms enable global information sharing, creating vast sensing networks for rapid data collection and analysis.
• The COVID-19 pandemic spurred extensive social media mining to address public health challenges from diverse angles.

Purpose of the Study:

• To review and categorize social media data mining studies conducted within the COVID-19 context.
• To document COVID-19 social media data archives for research reproducibility.
• To discuss challenges and future directions in social media analytics for public health.

Main Methods:

• Systematic review of social media data mining studies related to COVID-19.
• Categorization of studies into six domains: early warning, human mobility, communication, public attitudes, infodemic, and hatred/violence.
• Documentation of publicly available COVID-19 social media datasets.

Main Results:

• Social media mining efforts during COVID-19 spanned six key domains, offering insights into various aspects of the pandemic.
• Identified essential features of publicly available COVID-19 social media data archives to support research.
• Highlighted seven challenges in social media analytics impacting COVID-19 findings.

Conclusions:

• Social media mining is a valuable tool for understanding and responding to public health emergencies.
• Addressing challenges in social media analytics is crucial for reliable COVID-19 research.
• Future research directions can leverage social media data for enhanced public health surveillance and response.