Research On The Influence Of Social Media Communication Based On Community Structure

With the continuous development of the Internet,social networks have become the primary channel for netizens to share information and knowledge.The dissemination of information in social networks greatly facilitates the promotion of new technologies and new ideas.In the process of interaction,a person can belong to multiple groups at the same time.As a hub node between communities,this person will play a very key role in the dissemination of information,and then the network topology will change.Therefore,the influence analysis based on community structure plays a vital role in understanding the behavior characteristics of nodes,revealing the network communication dynamics and analyzing the network topology.The following are the main contributions of this paper:(1)The idea of overlapping communities is introduced into the influence maximization of social networks,and the IMTOC(influence maximization through overlapping communities)algorithm is put forward,and the public real social network data set is used to calculate the influence and obtain the seed nodes.Compared with other algorithms,IMtoc algorithm can obtain a larger node influence range with shorter running time.(2)An IOCI(Identification of Overlapping Community Influence)algorithm is proposed to rank the influence of overlapping communities.Through the idea of reorganization,the influence maximization algorithm is extended to the community level,and it can rank according to the influence of overlapping communities.The experimental results show that this method retains the characteristics of the number of community members;And for sparse networks,the influence of the community can also be judged when there is no obvious difference in the number of members of each community after community division.(3)Using Anybeat,an open social network data set,the network is displayed visually by Gephi software according to the relevant algorithm proposed in this paper.According to the analysis of data and network diagram,the characteristics of community structure in social networks are revealed,and the rationality and accuracy of the algorithm given in this paper are confirmed.In this paper,the previous influence analysis is improved and expanded from individual to local.The proposed algorithm can not only find out the most influential nodes,but also apply the classical influence algorithm to the community level to find out the communities with high influence.Finally,combined with visual analysis,it can clearly understand the status of influential nodes in the network and has high application value.