Modeling And Experimental Analysis Of Social Network Based On Natural Growth Model And Reconstruction Model

Aiming at constructing a reasonable social network structure model,social network structure modeling is the basis of many other research fields of social network.However,because of privacy protection and so on,it is almost impossible to obtain all the data needed to build the model.Although there are many methods which use incomplete data to construct the social network structure model,most of them have some problems.For example,some static models can not reflect all the topological characteristics of the social network,and the nodes grow infinitely over time in some of the dynamic models.In this thesis,technology needs are put forward based on the analyses of the application scenarios.Then three social network structure models follow on,which are the natural growth model based on historical events,the reconstruction model based on local network information and the time-space fusion social network model.The three models above are designed from different perspectives of time,space and time-space fusion.And the first one is based on the general behavior principles and the influence mechanisms of events on network structure.The second one revises the former common neighbors and node attributes similarity indexes.Many basic theories,such as the natural selection of social network are applied in the last model.Based on the Netlogo software platform,using different data and methods,the experiments of the three models above are carried out.The results of them show that the three network models have good accuracy under different conditions individually.Therefore,users can choose different models to build the social network structure according to the different data acquired in the specific application scenarios.In addition,due to the changes of the external environment and the state of the nodes,the structure of the social network changes constantly.However,this change of the social network always makes the social network structure more stable on the whole.