Mining And Forwarding Prediction Of Key Nodes In Microblog Information Dissemination

Online social networks have grown rapidly in recent years,and online social platforms and media functions are more detailed.However,due to the complexity of the factors affecting the dissemination of information,it is become hot and difficult problems that found the key nodes that influence the dissemination of information and the further prediction of the dissemination of information.Social influence describes the size of the influence that user in the social platform impact others,social influence of the user in the process of information dissemination often play an important role.Forward prediction based on the existing form of information dissemination,predict the next period of time which users will participate in the dissemination of information in the process.Most of the existing forwarding prediction algorithms are studied from the macroscopic point of view,and the diffusion rate of information propagation is analyzed by probabilities and thresholds,and no specific nodes are analyzed and studied.For the above reasons,this paper studies the key nodes and forward forecasting problems in the process of social network information dissemination.The specific work is as follows:(1)In the social network,the information dissemination has obvious concatenation relationship.In order to understand the specific communication mechanism of the information,after the information is transmitted,the topology of information dissemination is constructed and the information dissemination process is analyzed.(2)Based on the analysis of PageRank algorithm,the NodeRank algorithm is designed to make it more suitable for the complex network structure composed of information dissemination,and the influence of each node participating in information dissemination is quantified to find the key nodes in the process of information dissemination.(3)The key node in the process of information dissemination is an important factor influencing the information dissemination mechanism,which is the main driving force for the large-scale and rapid dissemination of information.This paper analyzes the factors that affect the user's participation in the information forwarding,divides these factors into four categories:text feature,user characteristic,interaction feature and time characteristic,and designs the quantization method of some features.(4)use the key node as the source of information dissemination,predict the user's forwarding behavior.According to the existing forwarding node,the SVM classification model is used to classify the nodes that have not yet participated in the information dissemination.(5)On the basis of the real data set,the comparison experiment of key node mining and information forwarding prediction is carried out,and the experimental results are analyzed.This paper analyzes the two problems:key node mining and user forwarding prediction in information communication process.In the problem of key node mining,we mainly solve the problem of how to measure the social influence of nodes.On the issue of user forwarding forecasting,we use the key nodes as the source node of information dissemination,and use SVM classification model to predict the nodes of the source node.Whether the interaction history node will participate in the process of information dissemination.