Location
Hilton Waikoloa Village, Hawaii
Event Website
https://hicss.hawaii.edu/
Start Date
7-1-2025 12:00 AM
End Date
10-1-2025 12:00 AM
Description
To understand and characterize the diffusion trends of opposing viewpoints on Twitter, we applied two epidemiological models to six datasets related to COVID-19. We compared the results of the SIR (Susceptible, Infected, Recovered) and the SEIZ (Susceptible, Exposed, Infected, Skeptics) epidemiological models. We collected six datasets indicative of polarizing viewpoints related to contentious subjects surrounding the COVID-19 pandemic. Three of the datasets fall into an anti-subject hashtag group, and three fall into a pro-subject hashtag group. The timeframe of each dataset is from January 1, 2020, to the end of 2021. Our findings demonstrate that while both the SIR and the SEIZ models can evaluate the propagation trends of the polarizing viewpoints, the SEIZ model is more accurate with relatively less error compared to the SIR model. This work sets the stage for ultimately leading to the ability to develop methods to prevent the propagation of ideas that lack scientific evidence while promoting the spread of scientifically backed ideas.
Recommended Citation
Maleki, Maryam and Agarwal, Nitin, "A Comparative Evaluation of the SIR and SEIZ Epidemiological Models to Describe the Diffusion Characteristics of COVID-19 Polarizing Viewpoints on Online Social Networks" (2025). Hawaii International Conference on System Sciences 2025 (HICSS-58). 4.
https://aisel.aisnet.org/hicss-58/dsm/data_analytics/4
A Comparative Evaluation of the SIR and SEIZ Epidemiological Models to Describe the Diffusion Characteristics of COVID-19 Polarizing Viewpoints on Online Social Networks
Hilton Waikoloa Village, Hawaii
To understand and characterize the diffusion trends of opposing viewpoints on Twitter, we applied two epidemiological models to six datasets related to COVID-19. We compared the results of the SIR (Susceptible, Infected, Recovered) and the SEIZ (Susceptible, Exposed, Infected, Skeptics) epidemiological models. We collected six datasets indicative of polarizing viewpoints related to contentious subjects surrounding the COVID-19 pandemic. Three of the datasets fall into an anti-subject hashtag group, and three fall into a pro-subject hashtag group. The timeframe of each dataset is from January 1, 2020, to the end of 2021. Our findings demonstrate that while both the SIR and the SEIZ models can evaluate the propagation trends of the polarizing viewpoints, the SEIZ model is more accurate with relatively less error compared to the SIR model. This work sets the stage for ultimately leading to the ability to develop methods to prevent the propagation of ideas that lack scientific evidence while promoting the spread of scientifically backed ideas.
https://aisel.aisnet.org/hicss-58/dsm/data_analytics/4