Research On The Dynamics Of Microblog Propagation Based On Complex Network

Complex network is a new method for complex systems study in natural science.Microblog is a social media platform allowing users to share real-time informationthrough the user relationship network which are established by the mechanism of“following”. Messages are spread through the user relationship network by users’behavior of “forwarding”. In this paper, we mainly apply complex network analysis tothe study of transmission dynamics on microblog message spread.The characteristics of microblog message propagation are instant, convenient andpublic. With the growing number of microblog users, especially with the explosiveway of spreading message on microblog, the message can be spread quickly over themicroblog network, which makes the security issues more and more prominent. Anysensitive social issue on microblog can be easily rendered and transmitted by users formalicious dissemination, which then causes adverse public opinions. Therefore, achallenging task is to describe the propagation process of microblog message, minethe propagation characteristics of microblog, and then take efficient measures tocontrol message spread.In this paper, the transmission dynamics of microblog message propagation isinvestigated from the following three aspects: firstly, the microblog propagationnetwork are established from the data of Sina microblog, the topological structurecharacteristics of the network are analyzed with complex network analysis methods,and the results reveal the propagation rule and characteristics of microblog; secondly,according to the propagation rule of microblog, an algorithm is proposed to establishthe microblog propagation network, and its degree distribution model is derived basedon the mean-field theory; thirdly, the control effects of six kinds of immunizationstrategy are analyzed based on node centrality. As experimental results show, thedegree distribution of microblog propagation network is a time-dependent model, andit is a power-law distribution at specific time. Immunizations at the center node andthe degree centrality nodes are the most effective microblog controlling strategy, itleads to rapid drop-off of the microblog’s influence scale.