Research On Individual Behavior Prediction Models In Social Networks

With the development of healthy social network and the improvement of people's living standard,it is urgent to study the prediction of individual behavior.Therefore,more and more scholars begin to pay attention to the prediction of individual behavior in healthy social network.In previous studies,many scholars have studied and discussed the characteristics of sports,movement propagation and behavior prediction in healthy social network.However,in these studies,time sensitivity and social impact are ignored.On this basis,this paper discusses the problem of individual behavior prediction in the healthy social network,which is mainly reflected in the following three aspects:First of all,considering the time sensitivity of individual behavior fusion time to reduce social network function prediction problem,put forward the TSGP model,the model incorporates individual factors,social factors and time sensitive factors based on the basic model of Gauss,but Gauss model only consider the ordinary individual factors,we in order to capture the social influence,Gauss will extend the basic processing model.Experimental verification on real data sets and synthetic data shows that the TSGP model proposed by us has better prediction accuracy.Secondly,the diversity,dynamics and hidden social impact of user behavior make the problem of behavior forecast a more serious challenge.In order to solve this problem,we study the basic algorithm of RBM in depth,StRBM model is proposed,the model will be the social impact of the individual factors,obvious and social influence hidden together,formed a historical layer,visible layer and hidden layer,using the parameter settings in the dynamic bias the three layer together.Finally,the validity of the proposed model is verified by experiments on real data sets and synthetic data sets.Finally,the study of social communication is how to spread the level of people's activities.Based on the IC model and CSI model,the CSP model of this paper is put forward.It uses this model to analyze the spread of sports activities and their social impact,and divide them into influencers,affected persons and unaffected ones.Experiments on real data sets demonstrate the validity of the proposed model and verify the correlation between the three social groups.