Research Of Coupled Dynamics Of Message Propagation And Opinion Formation On Two-layer Network

Research on the dynamics of complex networks is one of the key focuses of network science,including phenomena such as percolation,synchronization,opinion formation,epidemics,and information spreading.These studies help to reveal the laws and mechanisms behind human society and biological evolution,and provide scientific guidance for effectively controlling the evolution of complex systems.In the early stages of the rise of network science,researchers studied isolated dynamics processes on single complex systems and achieved good results.However,isolated dynamic processes cannot fully describe the various dynamic phenomena in actual complex systems.In fact,the vast majority of complex systems are composed of multiple interacting subsystems.These complex systems can be described by multilayer networks,such as multilayer social networks that interact with multiple social platforms and multilayer transportation networks that couple multiple transportation modes.In multilayer networks,the dynamic processes of different network layers interact with each other.For example,there may be cross-immunity or mutual promotion effects of different infectious diseases during the spread of the diseases,and there is a mutual synergy between synchronization processes in the brain’s neural system and nutrient transport processes.There is also an asymmetric coupling effect of inhibition-promotion between the spreading of information about epidemics and the spread of epidemics themselves.The study of coupled propagation dynamics processes on multilayer networks is crucial,as it helps deepen our understanding of real complex systems and predict and control them effectively.The formation of group opinions has always been a topic of great interest in research.In modern democratic societies,major decisions often reflect the consensus of group opinions.Finding suitable models to describe the process of group opinion formation has important practical significance.However,the process of opinion formation among crowds is often not independent and can be influenced by other factors,such as the spread of message on social media.Given that the time scale of group opinion formation in reality is much longer than that of message propagation on social media,and differences of evolution scenarios between two dynamic processes and the coupling effects between them.The first part of this thesis establishes a noisethreshold voter-SIR coupled propagation model based on a double-layer coupled network that reflects the difference in the time scales of the two dynamic evolutions and studies the influence of interlayer coupling on the co-evolution of propagation dynamics.Different theoretical methods are established to analyze the model for homogeneous and heterogeneous networks.The theoretical analysis and Monte Carlo simulation results both indicate that the cooperative coupling between the two dynamic processes promotes the emergence of the bistable phase and hysteresis phenomenon.Interestingly,the model exhibits a phase flipping phenomenon in small systems,where group opinions switch back and forth between two consensus states.The study shows that the fluctuation of individual opinion flipping strength is the cause of this phenomenon.This study illustrates the important influence of message propagation on social media on group opinion consensus,and the necessity of a certain proportion of dissenters for group opinion flipping in the case of limited resources for social network propagation.This research has profound implications for controlling public opinion and refuting rumor.In general,the time scale of opinion formation in a group can be considered much larger than the time scale of message propagation on social media.However,this assumption does not hold in certain extreme scenarios.In recent years,research has shown that the relative time scale between the two dynamics has a significant impact on their coupled evolutionary process.Building on the first part of the study,the second part of this thesis establishes a noise threshold voter-UAU coupling propagation model based on a double-layer network with an adjustable relative time scale,and investigates the impact of the relative time scale between the two dynamics on their co-evolutionary dynamics.The results show that a smaller time scale for message propagation compared to opinion formation is more conducive to the formation of bistable state,due to the differences in the evolutionary characteristics of the two dynamics and their collaborative interactions.Specifically,when the overall opinion of the group is in different states,the relative time scale between the two dynamics has the opposite effect on the formation of a consensus for a certain opinion.This work fills a research gap on the impact of relative time scale on cooperative coupled dynamics and deepens our understanding of the influence of relative time scale on the evolution of coupled dynamics.