The Reaearch Of The Effect Of Individual Behavior On Epidemic Spread And Random Walk Dynamics In Time-varying Networks

The development of network science brings new perspectives for the study of various real systems in real life.In recent years,with the rapid development of Internet technology,the acquisition of large amounts of time series data provides convenience for the study of dynamic behavior on the network.Among them,network propagation and random walk are one of the hot topics in network science research.It is a prerequisite for controlling the spread of disease to study the transmission behavior on the network and master its transmission law.In addition,the study of random walk process on the network can provide a basis for efficient immunization strategies.Due to the huge scale of real network system,the complexity of individual behavior and the changeability of network structure,it is faced with many challenges to propose effective immunization strategies.This dissertation focuses on the behavior patterns of individuals,studies the evolutionary mechanism of sequential networks,and explores the influence of individual behavior patterns on the propagation behavior and random walk process on the network,so as to provide theoretical guidance for the design of effective immune strategies.In this dissertation,the following innovative results have been achieved in the study of communication,random walk and immunization strategies of temporal networks:1.Based on the individual's memory behavior,a kind of temporal network model is proposed.Based on the network model driven by individual's activity and attraction,a temporal network model with individual memory behavior is proposed.The influence of individual memory behavior on network topology and connection mode is studied,and the influence of individual memory on communication behavior is discussed.The experimental results show that individual memory can change the connection mode of network system and have different effects on different propagation models on temporal networks.For SIR epidemic model,individual memory has little effect on transmission threshold,but it will reduce the final infection ratio.However,for SIS epidemic model,when individual activity is negatively correlated with attractiveness,individual memory reduces transmission threshold and promotes epidemic transmission.Individual memory has little effect on the transmission threshold when they are independent and positively correlated with each other.2.Based on the individual's mutual selection behavior,a kind of temporal network model is proposed.Based on the temporal network model driven by individual activity,a kind of network model with individual mutual selection behavior is proposed,and the random walk process under different mutual selection modes is studied.Theoretical analysis and experimental results show that different mutual selection modes havedifferent effects on random walk process.The probability that a traveler stays on the node is not only related to the activity of the node,but also affected by the node's ability to accept connections.3.Based on the random walk mechanism,a kind of immune strategy for temporal networks is proposed.Based on the random walk process,the important nodes in the network are identified,and a kind of immune strategy based on the random walk process is proposed.The efficiency of the immune strategy is compared with that of the classical random immune strategy and the target immune strategy.The results showed that the immune strategy based on the random walk was better than that of under different immune proportions.When the immune proportions large,the immune strategy of random walk almost achieved the same immune effect as that of target immunity.Compared with the target immune strategy,the advantage of the immune strategy based on random walk is that it does not need to know the information of the nodes in the network beforehand,so it can be applied to the network system with incomplete information.