Modeling Of Information Transmission Dynamics And Immunization Strategies On Complex Networks

At present,the studies on complex network mainly include empirical analysis on network structure,network evolution and modeling,network information mining and prediction,network control,spreading dynamics behaviors modeling and immunization on network.Among these studies,the spreading behaviors modeling and immunization on complex network,such as rumor propagation and control,epidemic spreading and contol,computer virus spreading and control,the prevention and control of cascading failure on electric power network and so on,belong to the application problems of complex system theory and have become the research hotspots in many fields.In this dissertation,the information spreading behaviors modeling and immunization on complex network are deeply investigated with consideration of the influencing factors of spreading behaviors including network topological structure,individual behavior and external information.On the basis of interactive Markov chains theory and mean field theory,three new spreading models on complex network are established,and their dynamics characteristics including sprading threshold and final scale of spreading are analyzed.Additionally,two improved immunization strategies are proposed for prevention and control of malicious information spreading.Four main contributions of the dissertation are as follows:1.On the basis of mean field theory,a modified susceptible-exposed-infected-removed(SEIR)rumor spreading model is proposed by considering hesitating mechanism and the attractiveness and fuzziness of the content of rumors.Accordingly,mean field equations are formulated to characterize the dynamics of SEIR model on both homogeneous and heterogeneous networks.Then a steady-state analysis is conducted to investigate the spreading threshold and the final rumor size.Simulations on both artificial and social networks show that the speed of rumor spreading obeys the relation“heterogeneous network>homogeneous network”,whereas the final scale of spreading obeys the opposite relation.In addition,a decrease of fuzziness can effectively increase the spreading threshold of the SEIR model and reduce the maximum rumor influence.2.A novel two-stage susceptible-infected-authoritative-removed(SIAR)rumor spreading model for complex homogeneous and heterogeneous networks is proposed by using interactive Markov chains method.The differential dynamic equations of SIAR model are derived to describe the dynamical interaction between the rumors and authoritative information.Monte Carlo simulations on the same size of WS and BA networks are performed to characterize the dynamic interaction process of network rumors and authoritative information,and a new critical value ?_c' which has an extremely different meaning with the traditional spreading threshold ?_c is defined.Simulation results demonstrate that the spreading threshold ?_c of SIAR model on both WS network and BA network is independent of lag time ? of authoritative information,while the new critical value ?_c' is related to the lag time ?.With the decrease of ?,the ?_c' increases.Correspondingly,when there is a rumor spreading in real world,the sooner the authoritative organizations or media releases the authoritative information,the less negative impact the rumors will bring.In addition,the increase of the diffusion rate of authoritative information in homogeneous networks is able to decrease the final rumor influence more effectively than that in heterogeneous networks.3.Considering dynamical disease spreading network consisting of moving individuals,a new double-layer network is proposed,one where the information diffusion process takes place and the other where the dynamics of disease spreading evolves.On the basis of Markov chains theory,a new model characterizing the coupled dynamics between information diffusion and disease spreading in populations of moving agents is presented and corresponding state probability equations are formulated to describe the probability in each state of every node at each moment.Monte Carlo simulations are performed to characterize the interaction process between information and disease spreading and investigate factors that affect spreading dynmics.Simulation results show that the increasing of information spreading rate can reduce the scale of epidemic spreading in some degree.Shortening disease period and strengthening consciousness for self-protection by decreasing healthy individual's range of motion both can effectively reduce the final refractory density for the disease,but have less effect on the information spreading.In addition,the increasing of vaccination rate or decreasing of long-distance jumps can also reduce the scale of epidemic spreading.4.An improved dispersion-based immunization strategy with consideration of node's degree and clustering is proposed for the immunization of scale-free networks with high degree of clustering.The main aim of the dispersion-based strategy is to iteratively immunize the nodes which have a high connectivity and a low clustering coefficient.Effectiveness of the proposed strategy is verified through comparing with two typical local immunization strategies:clusterrank-based immunization and degree-based immunization on both real-world and scale-free networks with a high level of clustering.Monte Carlo simulations show that the dispersion-based strategy requires small computing burden compared with the clusterrank-based immunization,and the performance of dispersion-based strategy is superior to that of two typical strategies.In addition,three improved global immunization strategies based on two rounds of selection are proposed to determine a fixed number of immunized nodes.The first-round selection is determined by the ranking criterion of degree centrality and the second-round selection based on the first-round results is determined by the ranking criterion of another node attribute,such as clustering coefficient,betweenness and closeness.Simulation results demonstrate that the presented strategies based on two-rounds of sorting are effective for heterogeneous networks and their immunization effects are better than that of the high degree and the high degree adaptive immunizations.