Research On Weibo Public Opinion Communication Based On Scale-free Network With High Clustering Coefficient

With the development of Web2.0,the media of public opinion has changed greatly.Weibo has become the main place of public opinion fermentation in the mobile social platform due to its large number of users,weak link relationship,real-time topic update and other characteristics.This paper constructs a complex network of weibo users' follow relationship as the base map of public opinion communication model.At the same time,an improved SEIR model is proposed.With the help of netlogo,multi-agent simulation modeling is completed to simulate the process of public opinion communication.This paper proposes an improved triangle connection algorithm to build a complex network of weibo users' follow relationship.The simulation system not only conforms to the scale-free characteristic of power-law distribution,but also conforms to the small world characteristic of high aggregation coefficient and low average distance.At the same time,there is a certain random connection.Then based on the classic SEIR model,an improved SEIR model is proposed.It is proposed that the transition probability between states is not a fixed value,but a dynamic change with time and the number of times that people have heard of public opinion.Because of the instantaneity of information,the propagating state is instantaneous,and the next moment is to return to the known state.Through the above ideas to design the model,compare the model simulation data with the crawled real weibo forwarding data,and determine the feasibility of the model.With the help of the above model,this paper explores the influence of transition probability between different states on public opinion communication and draws some conclusions,including: the proportion of nodes in known states has a great influence on the scope and time of public opinion communication,the larger the proportion of nodes in known states is,the longer the time of public opinion communication is;the greater the probability of conversion from known states to senseless States is,the shorter the time of public opinion communication is;and The larger the probability transmission rate is,the wider the public opinion transmission range is,and the time increases first and then decreases;when the senseless state and the communication state are both large,the public opinion will spread in a short time and a small range;the larger the hot search coverage rate is,the longer the public opinion transmission range is,the shorter the time is,but the senseless rate and the transmission ratedetermine the existence time of the hot search;a large number of nodes with small output are at the same time As the initial communication node,it can also produce the same effect of public opinion communication with the influential initial communication node in the same time.Finally,this paper also makes a theoretical analysis of the real weibo public opinion control methods,and explains the application of different methods.Some suggestions are put forward for the guidance and control of public opinion.