Location
Online
Event Website
https://hicss.hawaii.edu/
Start Date
3-1-2023 12:00 AM
End Date
7-1-2023 12:00 AM
Description
The activity of bots can influence the opinions and behavior of people, especially within the political landscape where hot-button issues are debated. To evaluate the bot presence among the propagation trends of opposing politically-charged viewpoints on Twitter, we collected a comprehensive set of hashtags related to COVID-19. We then applied both the SIR (Susceptible, Infected, Recovered) and the SEIZ (Susceptible, Exposed, Infected, Skeptics) epidemiological models to three different dataset states including, total tweets in a dataset, tweets by bots, and tweets by humans. Our results show the ability of both models to model the diffusion of opposing viewpoints on Twitter, with the SEIZ model outperforming the SIR. Additionally, although our results show that both models can model the diffusion of information spread by bots with some difficulty, the SEIZ model outperforms. Our analysis also reveals that the magnitude of the bot-induced diffusion of this type of information varies by subject.
Recommended Citation
Maleki, Maryam; Mead, Esther; Arani, Mohammad; and Agarwal, Nitin, "Measuring the Interference Effect of Bots in Disseminating Opposing Viewpoints Related to COVID-19 on Twitter Using Epidemiological Modeling" (2023). Hawaii International Conference on System Sciences 2023 (HICSS-56). 3.
https://aisel.aisnet.org/hicss-56/dsm/digital_methods/3
Measuring the Interference Effect of Bots in Disseminating Opposing Viewpoints Related to COVID-19 on Twitter Using Epidemiological Modeling
Online
The activity of bots can influence the opinions and behavior of people, especially within the political landscape where hot-button issues are debated. To evaluate the bot presence among the propagation trends of opposing politically-charged viewpoints on Twitter, we collected a comprehensive set of hashtags related to COVID-19. We then applied both the SIR (Susceptible, Infected, Recovered) and the SEIZ (Susceptible, Exposed, Infected, Skeptics) epidemiological models to three different dataset states including, total tweets in a dataset, tweets by bots, and tweets by humans. Our results show the ability of both models to model the diffusion of opposing viewpoints on Twitter, with the SEIZ model outperforming the SIR. Additionally, although our results show that both models can model the diffusion of information spread by bots with some difficulty, the SEIZ model outperforms. Our analysis also reveals that the magnitude of the bot-induced diffusion of this type of information varies by subject.
https://aisel.aisnet.org/hicss-56/dsm/digital_methods/3