We study the spread of a simulated epidemic in a network of individuals who may either contract a disease through sexual contact with an infected nearest neighbor or use safe sex practices under the influence of neighbors who are already adopting precautions. We show that both interaction between susceptible and infected individuals and the imitation of opinions concerning safe sex practices between individuals in favor of using such practices and those opposed to them leads to a phase transition. If the parameters of the epidemic are in the supercritical state, corresponding to an unlimited growth of infection, the interaction parameter of the sociological debate must also be in the supercritical state to control the spread of infection, and bring the system to criticality.
Adopting a theoretical perspective like that of multilayer complex networks, we study the case where the epidemic network is under the influence of the above-mentioned sociological debate. We show that at criticality this debate generates clusters of individuals in favor of safe sex practices and clusters of individuals opposing their use. We study the influence of a sociological debate on whether to use safe sex or not, on the spreading of sexually transmitted infections. We show that due to this debate in the epidemic network a pattern mirroring the structures of the sociological network appears.
Finally, we introduce a feedback of the epidemic network on the sociological network and prove that due to this feedback the sociological system undergoes a process of self-organization keeping it at criticality. We hope that these results have the effect of giving interesting suggestions to behavioral psychologists and information scientists actively involved in the analysis of the social debate on the moral issues connected to sexual activities.
Analysis of the susceptible-infected-susceptible epidemic dynamics in networks via the non-backtracking matrix
本文研究了任意结构的有限有向加权网络上的随机传染病易感模型。我们提出了一个新的下界的指数率,在该概率的节点被感染的衰减随着时间的推移。这个界限直接与依赖于网络的非回溯矩阵和关联矩阵的矩阵的前导特征值有关。这个矩阵的维数是N+M,其中N和M分别是节点和边的数量。我们表明,这个新的下限改善了现有的所谓的淬火平均场论的约束。虽然从最近开发的二阶矩闭合技术获得的界限需要计算N^2 x N^2矩阵的前导特征值,但我们在数值模拟中说明了新的界限更紧,同时对于稀疏网络计算成本更低。我们还提出了相应的传染病阈值的表达式的线图的邻接矩阵和给定的网络的非回溯矩阵。
Impacts of bridging nodes on the epidemic activation mechanisms
Universidade Federal de Viçosa(维索萨联邦大学);Universidade de São Paulo(圣保罗大学);Instituto Federal Fluminense(弗鲁米嫩塞联邦学院);National Institute of Science and Technology for Complex Systems(复杂系统科学与技术国家研究所)