Research Of Virus Propagation On Micro-blog Online Social Network

With the rapid development of online social networks, the transmission of information becomes more and more quickly, and the costs fall sharply. Meanwhile, the spread of the virus has caused huge loss to the society. Therefore, for curbing the virus and reducing the loss, it is essential to study the mechanism and to understand the rules of the virus propagation on online social networks.In recent years, researchers have made great achievements on virus propagation in the interpersonal network and Internet with theory of complex networks. Micro-blog networks have more complex characteristics comparing with the traditional complex network, and the traffic, dynamic evolution of the micro-blog network play a non-negligible role in the process of virus propagation. In order to sina micro-blog as research object, this paper presents a novel model of virus propagation, ISHR, based on the existing research and analyzes the influence to the virus propagation on dynamic evolution. The main works are listed as follows:(1) In order to analyze the basic topological structure of complex network, this paper adopts a crawler to obtain the micro-blog user relational data, and establish the model of network. With the help of Pajek and Matlab software, we analyze the static characteristics of the network, including the degree distribution, the shortest path length distribution, the clustering coefficient. The results above show that micro-blog online social network has small world properties, scale-free, and community structure.(2) Considering the communication flow, the model of virus propagation ISHR is proposed based on the traditional model of virus propagation. The results of the simulation with Matlab indicate that the possibility of outbreak of the virus increase with the transmission of information packets between the users. In addition, under a certain amount of packets, the degree of the infection on virus significantly reduces with the increasing of immunization rates. In contrary, the degree of infection large rises with the increasing of the effective spreading rate.(3) Among the characteristics on dynamic evolution of the micro-blog network, this paper analyzes the influence of the network module, bridge node, and power index to the virus propagation. The results show that the spread of the virus could be suppressed while the network has obvious community structure. However, the smaller degree of bridge node or the larger of the power-law index could promote the virus propagation.In addition, this paper puts forward the improved acquaintance immunization strategy based on the betweenness of node. The results show that the improved acquaintance immunization strategy could achieve better effect than the classic acquaintance immunization strategy. Based on the obtained results, people could easily understand the law of virus propagation and implement immunization strategy more effectively.