Researchers receive NSF RAPID award to understand how individuals are socially influenced online during COVID-19

For most people, social media is Instagram selfies, LinkedIn articles and losing yourself in the world of Facebook. For Arif Mohaimin Sadri and M. Hadi Amini, social media means valuable data to analyze the public response to crises like COVID-19. 

Sadri is an assistant professor in the Moss Department of Construction Management at the Moss School of Construction, Infrastructure & Sustainability. Amini is an assistant professor at the School of Computing and Information Sciences (SCIS). The Principal Investigators (PIs), Sadri and Amini, received a grant from the National Science Foundation as part of NSF’s Rapid Response Research (RAPID) program designed to aid in the modeling and understanding of the spread of COVID-19.

The focus of the $79,380 grant—titled “#COVID-19: Understanding Community Response in the Emergence and Spread of Novel Coronavirus through Health Risk Communications in Socio-Technical Systems”—will be exploring how individuals are socially influenced online, while communicating risk and interacting in their respective communities as the disease continues to spread.

The findings from the research will be useful for public health and emergency management agencies.

“We are collecting social media and crowd-sourced data to uncover how COVID-19 risk information is communicated through social media and how it influences the way people respond,” said Sadri. “This is an unprecedented opportunity.”

The researchers are aware this is time-sensitive data and every piece of information has to be verified.

“There’s a lot we can uncover in a short period of time,” said Amini. “We also need to verify the correctness of the collected data from social media before deploying it for decision-making.”

The two professors started monitoring the spread of the disease in January. They were able to use data-driven methods to monitor Twitter data, looking for the ways users are influenced based on what they see on social media leading to patterns of health-risk communication and community response. This encompasses how social media influences the protective actions taken by the public, such as self-isolation and social distancing.

The decision to use Twitter over other social media platforms came because Twitter is a publicly available domain with much fewer privacy restrictions compared to Instagram and Facebook. On Twitter, data can be followed in real-time.

The research team previously used social media data to study the response and aftermath of tornadoes, hurricanes and other major natural disasters. So they’re prepared for the next steps.

“We already have credible data,” said Sadri. “We’re revisiting research questions to make them more feasible.”

The benefits of this research can be felt overseas, in areas where resources are limited. Examples of these locations could be Dhaka, Bangladesh, or New Delhi, India. The research is low-cost and can benefit countries that are most impacted by COVID-19. Both Amini and Sadri are international faculty, and the scope of their interests spans globally.

This project has been a collaborative effort, crossing cultural and academic boundaries, for both researchers.

“FIU provides us with a diverse environment that allows for working with students from minorities and underrepresented groups,” said Amini.

Sadri is the principal investigator on the grant and is the director of the Network-driven Human-centered Infrastructure Resilience & Sustainability (N-HIRS) Lab. N-HIRS’s research focuses on the critical interdependence between social and infrastructure networks and integrates human proactive decision-making components into the civil infrastructure management challenges.

Amini is a co-principal investigator on the grant and is the director of the Sustainability, Optimization and Learning for InterDependent Networks laboratory (SOLID lab). SOLID lab is an interdisciplinary research group bridging the gap between theory and the real-world. Amini and his research group develop highly efficient computational algorithms (including machine learning and optimization) to deal with large-scale optimization, learning, and ultimately decision-making problems for interdependent networks.

Both labs are housed within the College of Engineering & Computing.