基于社交关系的流行病传播与免疫机制研究

发布时间：2018-05-17 07:21

本文选题：流行病传播 + 社交关系　；参考：《兰州理工大学》2017年硕士论文

【摘要】：现实中,每个人都身处多个不同的社交关系网络,并且在这些社交网络中扮演着各种角色。各种各样的社交关系相互之间交叠的同时又彼此隔离,例如在某个人朋友圈内的人们之间并不都是相互认识;同事之间也可能彼此又是同学关系;同一个人的朋友和亲戚之间可能完全不认识。无论是消息还是传染病都可以通过这些错综复杂的关系网传播和流行。而包含有传染病信息的消息与该传染病同时在人群中传播时,对传染病的传播到底会产生什么样的影响呢?毫无疑问,研究伴有信息传播的流行病传播过程是十分必要的。研究这些复杂的社交关系及其上流行的消息对传染病传播的影响,要先研究社交关系之间的信息交流特点。真实网络中个体获得疾病消息后,会根据以往的染病经历做出相应的规避行为,在本文中称其为个体应激性规避行为。网络中社交消息传播必然会对流行病传播过程产生影响。基于这个特点,本文运用统计物理、运筹学和计算机仿真等方法着重研究由疾病消息引起的个体应激性规避行为和由社交网络交叠产生的个体多关系角色特点两方面对流行病传播的影响:(1)本文在单群体内部基于个体应激性规避行为,分别在WS小世界和BA无标度网络上改进了SIS流行病传播模型,并进行理论分析和实验仿真。结果表明,由疾病消息引起的个体应激性规避行为能够促使网络在整体层面上具有适应、抑制流行病传播的能力。这个现象说明,个体可以通过自我学习获得预防被感染的能力,同时,网络也能够通过这种个体的特殊能力获得对流行病的网络整体防控能力。这种网络整体层面上学习如何适应疾病并逐渐调节健壮自己的行为,对快速应对未来流行病爆发和有效控制流行病再传播具有十分积极的作用。(2)为了更好地论述复杂的社交关系对流行病传播的影响,本文中构建一个不同于传统多社团网络的叠加社交网络,同时提出了全新的MR-SIS流行病传播模型,并进行理论分析和试验仿真。结果表明,网络中子网络的组成结构能够对流行病的传播过程产生影响,尤其是叠加子网络中的节点,控制这些节点可以有效地抑制疫情的传播蔓延。在这个基础上,本文提出基于叠加子网的熟人免疫策略,实验结果也表明该策略是切实可行的。
[Abstract]:In reality, everyone lives in multiple social networks and plays a variety of roles in them. All kinds of social relationships overlap and separate from each other, for example, not all people in the circle of friends know each other; colleagues may also be classmates. Friends and relatives of the same person may not know each other at all. Both news and infectious diseases can be spread and spread through these intricate networks of relationships. What kind of impact will the information containing infectious diseases have on the transmission of infectious diseases when they are transmitted in the crowd at the same time? There is no doubt that it is necessary to study the process of epidemic transmission with information dissemination. To study the influence of these complicated social relationships and popular news on the transmission of infectious diseases, we should first study the characteristics of information exchange between social relationships. After getting the disease information, the individual in the real network will make the corresponding evasive behavior according to the past infection experience, which is called individual stress-evading behavior in this paper. The spread of social information in the network will inevitably have an impact on the epidemic spread process. Based on this characteristic, this paper uses statistical physics, Operational research and computer simulation methods focus on the impact of individual stress-evading behavior caused by disease messages and the characteristics of individual multi-relationship roles generated by overlapping social networks on epidemic transmission: 1) Group based on individual stress-induced circumvention behavior, The SIS epidemic propagation model is improved on WS small-world and BA scale-free networks, and the theoretical analysis and experimental simulation are carried out. The results show that the individual stress-evading behavior caused by disease information can make the network adapt to the epidemic at the whole level and inhibit the spread of the epidemic. This phenomenon shows that the individual can acquire the ability to prevent infection through self-learning, and the network can also acquire the overall network ability to prevent and control the epidemic through the special ability of the individual. The network as a whole learns how to adapt to disease and gradually modulate its own robust behavior. Have a very positive role in responding quickly to future outbreaks of epidemics and effectively controlling their re-transmission) in order to better address the impact of complex social relationships on the spread of epidemics, In this paper, a superimposed social network, which is different from the traditional multi-community network, is constructed, and a new MR-SIS epidemic transmission model is proposed, and the theoretical analysis and experimental simulation are carried out. The results show that the structure of the network neutron network can affect the epidemic spreading process, especially the nodes in the superimposed sub-network, and the control of these nodes can effectively inhibit the spread of the epidemic. On this basis, an acquaintance immune strategy based on superimposed subnets is proposed, and the experimental results show that the strategy is feasible.
【学位授予单位】：兰州理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：G206;O157.5

【参考文献】