Sns Community Network Structure Based On Graph Theory-centrality Analysis

With the rapid rise of network communities, the study and analysis of networkcommunity structure has gradually become an academic hot topic. The networkspatial structure presented by the interpersonal nodes and groups in SNS communitynetwork has many differences with the traditional interpersonal relationship andurban spatial structure system, which presents a series of new characteristics andlaws. But it still has a large gap with the communication distance "barrier-free" stageamong interpersonal nodes of ideal SNS community, the effects of traditionalgeographical space is still a factor that can’t be ignored. In the study, We find thatSNS community network spatial structure shows a significant and diverse centricitycharacteristics, this kind of significant centricity characteristics has become the mostimportant competitive location elements of SNS community network, so it’sespecially necessary to analyze the centrality characteristics of network spatialstructure in SNS communities.The introduction of graph theory method is an innovative point for complicatednetwork community spatial structure research. In order to reveal the similarities anddifferences between SNS community network structure and realistic communitynetwork structure, This paper analyzes the centrality characteristics of its networkspatial structure based on graph theory method. It discusses the SNS communitynetwork space structure of the centrality characteristics mainly through four indicatorsof interpersonal nodes’ center degree, network graphs’ degrees center potential,condensation degree and concentration coefficient, trying to construct virtual networkspatial structure system under different interpersonal nodes and groups in China.First, this paper choose sina-blog which has the most users in China as theresearch object SNS community spatial structure, and selected five representativecases. Then we summarized interpersonal nodes, group numbers and link frequencydegree among interpersonal nodes using EXCEl software for the selected five cases.Secondly, We analyzed interpersonal nodes and groups in SNS communitynetwork spatial structure using ArcGis software, and then drew network space relationship graphs for interpersonal nodes’ distribution of five cases respectively. weconducted quantitative analysis for nodes’ attractive degree with different levels fromthe empirical aspects.Additionally, we conducted centrality analysis on network space relationshipgraphs of five cases,which was based on graph theory again and depended on Spsssoftware. And calculated interpersonal nodes’ degrees center degrees with differentlevels, degrees center potential, condensation degree and concentration coefficient, toreveal the centricity characteristics of different level node and the overall networkstructure in interpersonal nodes’ network spatial structure in SNS community network.Finally, we sum up the centricity characteristics of the SNS community networkspatial structure, it shows that:(1) There exists strong and weak interpersonal nodes different power in SNScommunity network, its spatial distribution shows disequilibrium and dispersion,interpersonal nodes with different levels exist obvious shelter sex, and as aconsequence, edge nodes often across the secondary node and direct contact anddirectly link with master nodes in the process of obtaining information, so it appears aphenomenon of relatively low centricity the same as center nodes.(2) The centripetal tendency of the SNS community network space structure isnot obvious, The differences of the degrees center potential are relatively small inevery network graphs and the don’t have extreme maximum numbers, thus verifyingthe decentralized non-spatial characteristics of virtual network. But the geopoliticalfactors still play an important role in some degree.(3) The important nodes in SNS community network structure determines thenetwork structure’s vulnerability, while the importance of interpersonal nodesbelonging to some city shows coupling characteristics between the real and virtualspace. Obviously, it sees that SNS community provides cross-domain communicationspace, in which friends break through the limitations of reality geographical distance,and group structure gradually also breaks through limitations of the circle layerstructure, then make interpersonal nodes’ relationship webified.(4) Each main node with different information level has higher node density in its attractive radius, and to the nodes beside the radius, influence of interviewerschasing the different information, nodes’ links can produce alien situation withclustering phenomenon.This paper conducts centrality analysis of SNS community network spatialstructure using graph theory method, so that it can make up for the blank on SNScommunity network structure characteristics research. It also provides moretheoretical support for exploring emerging network community links from geographyperspective and promotes the further development of human geography in theinformation age.