Social topic influence recognition method and device based on ternary association graph

Abstract

The invention relates to the technical field of social network hot topic control, in particular to a social topic influence recognition method and device based on a ternary association graph, and themethod comprises the following steps: obtaining data; constructing a path-user binary association graph model according to the acquired data; constructing a user-domain division model; constructing apath-user-domain ternary association graph model; according to a cross scoring strategy, performing positive and negative iterative voting on the path-user-domain ternary association graph model to mine key element nodes propagated by the hot topics; and according to the key element node information, dynamically deducing a hot topic propagation process in a complete life cycle by utilizing a timeslicing method. According to the method, the hot topic influence identification convenience can be improved. A key path, participation users and a propagation field can be identified according to thefinal score vector of each element. Dynamic mining of network public opinion topics can be realized.

Description

technical field [0001] The present invention relates to the technical field of social network hot topic control, in particular to a method and device for identifying social topic influence based on a ternary association graph. Background technique [0002] With the prevalence of social networks, there are a lot of false, defamatory and illegal information in online social networks. Due to the rapidity of network dissemination, the huge number of network users and the vastness of the network field, it is difficult to obtain the spread and spread of these illegal information. Effective control has adversely affected the public discourse space on the Internet. Therefore, how to prevent criminals from using network resources to create rumors and ensure the authenticity of online network information dissemination, the work of identifying the influence of hot topics is particularly important and urgent. [0003] In the prior art, the research work on topic influence identificatio...

AI Technical Summary

Problems solved by technology

However, the current research ignores the influence of such overlapping users and cross-fields on the maximization of topic dissemination, weakens the driving force of such users on topic dissemination, and reduces the accuracy of determining key nodes.

[0005] 2. Neglecting the mutual influence relationship between elements in topic network communication

The process of topic dissemination will involve key elements such as paths, users, and domains. How to describe the mutual influence relationship between key elements is very effective for quickly identifying the influence of topics on Internet public opinion, while traditional research is relatively difficult to mine these elements. Independent, does not closely link these elements, lacks unity

[0006] 3. Ignored the analysis of the dissemination situation of the topic at different stages

At present, analysis methods based on node attributes, reasoning methods based on propagation models, and analysis methods based on combining user attributes and network structures simply dig out topic sources or key nodes, but do not describe how topics evolve from One user links to another, how to link from one realm to another, lack of dynamics

Images