Abstract
Under the impact of internet populism, internet violence, and other noises on the internet, medical elites, who have a professional background, did not intend to share their opinions on the internet. Thus, misinformation about health is increasingly prevalent. We roughly divided the users in social networks into ordinary users, medical elites, and super-influencers. In this paper, we propose a communication model of health information based on the improved Hegselmann-Krause (H-K) model. By conducting MATLAB-based simulation, the experimental results showed that network noise was an important factor that interfered with opinion propagation regarding health. The louder the noise is, the harder it is for health opinions within a group to reach a consensus. But even in a noisy environment, super-influencers could influence the overall cognition on public health in the social network fundamentally. When the super-influencers held positive opinions in public health, the medical elite keeping silent had a noise-tolerant effect on opinion communication in public health, and vice versa. Thus, three factors concerning noise control, the free information release of medical elites, and the positive position of super-influence are very important to form a virtuous information environment for public health.
Highlights
In recent years, social media platforms (e.g., Facebook, Twitter, and WeChat1) have been widely adopted for information communication
We proposed a revised H-K model, which adopts a new coefficient named noise to reflect the consequence of medical elite silence to health information opinion exchange in social networks
With Medical Elites and Ordinary Users Involved We introduced the medical elites based on the experiment of simple simulation, and the remaining constraints were kept unchanged
Summary
Social media platforms (e.g., Facebook, Twitter, and WeChat1) have been widely adopted for information communication. This paper argued that while studying the propagation of health information online, we should consider the interactive modes and restrictions between individuals and pay more attention to the behavioral changes of nodes with different identities, especially the elite silence, under the interference of network noise. (1) When there was neither any elite user nor super-influencer adjacent to node i, i.e., any node that involved in interactions must be an ordinary userj, and the interaction condition satisfied xi − xj ≤ εCij. based on the improved H-K model, the algorithm of opinion interactions among ordinary users should include the noise factor, that was:. The opinion convergence is reached when the disparity of opinions within the group≤ 0.01
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
