Structural Analysis Of Social Network Based On Complex Network Technology

Social network is a set of social individuals and the relationships between them, which generally composed by nodes and the connections between these nodes. Since the 21 st century, the research in complex network has made a great progress and social network has obvious characteristics of complex network, so more and more scholars are applying complex network theory to study the social relations of social network. In this paper, we mainly study undirected-unweighted social network and directed-unweighted social network. We use complex network theory to analyze the structure of social network. Respectively, we detect communities and evaluate the importance of nodes in QQ group social network and Email social network.We mainly have done some research in this paper as following:Firstly, we use QQ social network as an example to analyze the undirected social network structure. We respectively study the whole structure, detect communities and analyze the importance of nodes. In this paper, the calculation method of density, average distance, connectivity and cohesion in undirected social network are introduced in detail. The authors describe two typical algorithms of community detection: GN algorithm and FN algorithm. The calculation method of degree centrality, betweenness centrality and closeness centrality in undirected network are expounded minutely. Secondly, we apply Email social network as an example to study the directed social network structure. The paper respectively analyzes the whole structure, detect communities and evaluate the importance of nodes. Detailed introduction on the calculation method of density, average distance, connectivity and cohesion in directed social network is given. The authors make a comprehensive description of two different algorithms of community detection in directed social network: the improved FN algorithm and LinkRank. The paper gives a detailed introduction to the calculation method of degree centrality, betweenness centrality and closeness centrality in directed network.The innovative points lie in the following:Firstly, in order to maintain the communication environment of QQ group, we present a new service for recommending administrators in QQ social network based on community detection and nodes importance exploration. Secondly, we put forward an Email marketing service based on community detection and nodes importance exploration in Email social network. On the purpose of promoting the corporate image and their products’ information, this Email marketing service send E-mail to the most important users of each community. The company can achieve an optimal effect of Email promotion in a certain time and cost. Thirdly, this paper proposes a strategy to optimize the communication environment of online social media, which is through building the internet real-name system, guiding important users actively and developing key users.