Font Size: a A A

Research On The Social Network Analysis

Posted on:2012-06-05Degree:MasterType:Thesis
Country:ChinaCandidate:P ZhaoFull Text:PDF
GTID:2180330392452223Subject:Applied Mathematics
Abstract/Summary:PDF Full Text Request
Social network analysis is a standard and method to analyze the social relationship andstructure and its attribute using the methods of graph theory, matrix and so on. The socialnetwork analysis is not only applied to individual relationships in micro level, but also cananalyze the macroscopic social phenomenon, organization structure and community relations.Now, the social network analysis has been regarded as one of the most persuasive perspectives instructure analysis. In this paper, the method of social network analysis is studied from twoaspects: theory and application.In the theoretical part, the development and conception of social network are described first.Then the social network model of kronecker self-similarity is presented on the basis of originalscale-free model. The construction of model, proof of definition and theorem, and connectivityof the network are studied systematically. At last, on the basis of dynamic community mining,the cluster membership for a node at time t depends both on its relationship with other nodes attime t and on its cluster membership at time t-1.In the part of application, the global trade relationship is analyzed by social networkanalysis. Regarding the trade data among76countries which is acquired from Economic andtrade BBS, the social network structure of global trade relationship is analyzed from the aspectsof scale, density, small-world property, centrality and so on by using the social network analysissoftware of ucinet and the sample is clustered in two ways: overlapping and non-overlapping.Taking the amounts of cooperation projects between developed countries and developingcountries and emission reductions as analysis sample, through the method of2-moulde networkanalysis, the sample data are analyzed in two aspects: from2-moulde network to1-mouldenetwork and from2-moulde to2department network to make the CDM have a visual display. Inthe end, the relationship of journal citation is analyzed by dynamic network analysis. Taking thedata of reference relationship between the authors of324papers which themes are “servicemanagement” in the core journal of China knowledge network from2007to2010and theauthors of corresponding citation literatures as analysis sample, the4years data is analyzed froma dynamic perspective.
Keywords/Search Tags:social network analysis, Community mining, cluster, 2-mould network, ucinet
PDF Full Text Request
Related items