Multi-Stage Information Spreading Model in Simplicial Complexes Driven by Spatiotemporal Evolution of Public Health Emergency

Abstract

The long cycle and cross-region evolution of the public health emergency have caused sharp turbulence in the society. Especially, the multi-stage and multi-channel information spreading triggered by the public health emergency has further aggravated the social panic, and even caused a chain reaction to cause greater chaos. In this paper, we extend the information spreading to simplicial complexes, and analyze the effect of the spatiotemporal evolution of public health emergency on individuals' behaviors at each stage by taking 2-simplex as an example, firstly. Then the multi-stage information spreading dynamics model is constructed and the basic reproduction number is estimated. What's more, empirical analysis and simulation are carried out to verify the effectiveness and analyze the dynamics of the model. The results show that the strength effect of 2-simplex interactions would expand information's maximum impact, and enlarge the spreading scale. Besides, the information's impact and spreading scale at the first stage are positively correlated with epidemic's initial impact. The information could spread rapidly or even rebound and produce new peaks when the public health emergency getting worsen. Especially, the rebound speed, peak and scale are negatively correlated with the initial impact of the public health emergency with epidemic local worsening. Moreover, the greater and the earlier of local and spatial deterioration, the greater the effect and the wider the information spreading scale. Furthermore, with the public health emergency in different regions getting worse or better in succession, the information's maximum impact and the spreading scale would be also enlarged. The results are useful to understand better the effect of public health emergency's spatiotemporal evolution on information spreading and suggest a promising way to weaken the negative effect of relevant information.