| Currently,in order to understand user's behaviors and to implement effective management of network message dissemination and diffusion,research on user's behaviors in Weibo Network has become popular.Analysis of Weibo user's behaviors can provide data support for Weibo public opinion guidance and Weibo rumors control.We proposes a behavioral dynamics model based on the interest decay,continuation habit and interaction impact on the Weibo user's behaviors.The improved method uses a statistical analysis of the potential laws of Weibo user's behaviors,extracts quantifiable internal driving factors,and then compares both based on previous research results and statistical analysis results.Through experimental verification,the improved analysis model can analyze the behaviors of Weibo users.The Research on User's behaviors in Weibo Network based on Human Dynamics runs in the real Weibo data,by crawling user's data,revising the model parameters and analyzing user's behavior to get the results.The correct rate reach to 0.9163.The final test result shows that the F value is 85% and the false detection rate is 8.5%,which shows a good discrimination effect on the Weibo user's behaviors analysis.An improved node importance evaluation method in social network is proposed named of ?NodeRank.The existing evaluation methods based on NodeRank for node importance in social network which ignore the effect of difference in the location of node on network structure,mainly focus on degree of node and weights of edge.The advantage of this improved algorithm is through thesis and experiment analysis to analyze,and the results show that the proposed method is more reasonable in evaluating node importance in social network.The research results of the dissertation are based on the whole life cycle of the development,dissemination,diffusion,and extinction of information on Weibo.This paper analyzes the information publishing behavior of Weibo users and evaluates their behaviors. |
PDF Full Text Request