Exploring Spreading Dynamics Of Message, Epidemics And Behavior On Complex Networks

In real social systems, the processes of spread and diffusion which can be widely observed are very important to the function of the organizations and institu-tions. Therefore the research in the dynamics of spread and diffusion present significant value both in theory and practice. For this purpose, the researchers in this field has developed serious of models and theories to study and describe various spreading phenomena, such as competition and diffusion of strategies, the development of trends and fads, the spread of epidemics, the propagations of new technologies and innovations, the spread of message and behaviors and so on. In this thesis, based on the elaborations of the definitions of complex net-works, evolutionary game theory and theory of spreading dynamics, we explore the evolution rules and propagations of cooperation by involving other mecha-nisms. Besides, we study the spreading of message and epidemics with certain mechanisms or in certain complex networks, respectively.Firstly, we review the most common parameters estimating network struc-ture like average degree, degree distribution and correlation, clustering coefficient, and typical networks such as regular lattice network, regular random network, random network, small-world network, scale-free network and multilayer net-work. Then we present the evolutionary game theory along with classical game models and introduce the concept of Nash equilibrium.In the second section, we propose a new agent-based model to capture the whole picture of the joint action of the memory effects, social reinforcement and decay effects in information spreading, by quantifying the complex conta-gion mechanisms as stickiness and persistence. The results renew our previous knowledge that messages can spread broader in networks with large clustering, which turns out to be only true when they can inform a non-zero fraction of the population in the limit of large system size. Then we investigate the epidemic spreading on onion network and find a positive correlation between oscillating behaviors and the hierarchies of the network. Third we extend our research to the metapopulation model to identify the different effects of three mobility modes on the outbreaks of epidemics.In the third section, we first combine the conformity with the prisoner’s dilemma game to identify the different effects of the minority rule and majority rule on the evolution of cooperation. Secondly, we propose a prisoner’s dilemma game model involving both altruistic and selfish punishment with avoiding mech-anism. Both pair approximation and numerical simulations show that not only the first-order social dilemma but also the second-order dilemma can be alleviated for certain values of punishment cost and fine.Finally, we make a summary and prospect of our thesis.