SIRA: a model for propagation and rumor control with epidemic spreading and immunization for healthcare 5.0

Kumar, Akshi ORCID: https://orcid.org/0000-0003-4263-7168, Aggarwal, Nipun and Kumar, Sanjay (2023) SIRA: a model for propagation and rumor control with epidemic spreading and immunization for healthcare 5.0. Soft Computing, 27 (7). pp. 4307-4320. ISSN 1432-7643

Preview

Accepted Version Download (681kB) | Preview

Abstract

Healthcare social networks have a significant role in providing connected and personalized healthcare environment with real-time capabilities. However, building resilient, robust and technology-driven healthcare 5.0 has its own barriers. Especially with social media's high susceptibility to rumors and fake news, these networks can harm the society. Many researchers have been investigating the process of information diffusion, and it has been one of the most intriguing issues in network analysis. Modeling rumor propagation is one of the prominent researched topics in recent years. Traditional models assume that rumor propagation happens only in one direction, where only supporters are supposed to be active, whereas, in a real-life situation, both supporters and deniers of the information operate simultaneously. In this paper, we introduce a model for the recovery of nodes in a setting where rumor propagation and rumor control happen simultaneously. We propose the Susceptible-Infected-Recovered-Anti-spreader model based on the notion of spreading of epidemics and also its applications to modeling the propagation of rumors and control of rumor. Our model assumes people have multiple forms of reactions to rumor, either posting it, deleting it or announcing the rumor as fake. This paper also suggests how the model can act as a simulation method to compare two node centrality algorithms where spreaders chosen from one centrality algorithm try to spread the rumor, and the anti-spreaders chosen from other centrality try to dispel the rumor and vice versa. We simulate the proposed algorithm on different weighted and unweighted real-world network datasets and establish that the experimental results agrees with the proposed model.