A method for tracking the evolution of public opinion based on a dynamic incremental probability graph model

Abstract

The invention discloses a public opinion evolution tracking method based on a dynamic incremental probability graph model, and the method comprises the steps: constructing an incremental random graphmodel, carrying out the efficient analysis of large-scale social network public opinion topics, tracking the evolution law of the topics, and predicting the future public opinion change. Specifically,the number of public opinion topics of a next time slice is dynamically determined based on the similarity-based correlation degree between relevant documents of each public opinion topic, so that the public opinion change condition of the next moment is predicted by reusing the posterior probability of historical public opinion information. According to the method, by combining historical publicopinion information, a public opinion evolution situation analysis method on time latitude is formed, and the evolution law of public opinion topics is found, so that future public opinions are predicted and controlled. According to the method, the public opinion topic number of the model can be determined more accurately, and the public opinion evolution process can be analyzed more accurately and efficiently.

Description

technical field [0001] The invention belongs to the technical field of big data analysis and application, and relates to the analysis of public opinion evolution, in particular to a method for tracking the evolution of public opinion based on a dynamic incremental probability graph model. Background technique [0002] Public opinion is the sum of various emotions, attitudes and opinions held by the public composed of various social groups in a certain social space on hot events, specific issues and social phenomena. Public opinion is generated around a specific topic, and the evolution of public opinion often changes with the development of the topic. Use the means of data analysis to discover changes in the topic to predict the development and changes of public opinion in the future, and track and analyze it to facilitate the prevention of emergencies. [0003] With the rapid growth of information, since the data is updated all the time, public opinion is constantly evolvi...

AI Technical Summary

Problems solved by technology

[0004] How to quickly know the evolution of public opinion is a current research hotspot. The problem with existing methods is that in the process of analyzing the evolution of public opinion using the probabilistic graphical model-LDA, each time you have to relearn, the efficiency and practice are relatively poor, and it will consume a lot of Time and the topic discovery has a delay, and the topic content cannot be presented to the user in time

And the choice of the number of LDA topics has a great impact on the interpretation of text data and the effect of topic recognition

The LDA model needs to manually set the number of topics, and there is a certain degree of blindness. If only the training model is used to continuously adjust the number of topics, it will undoubtedly consume a lot of time, and will affect the accuracy of topic discovery and the readability of content.

Images