University network public opinion prediction and analysis based on LogSparse Transformer model
ZHANG Youhai
Abstract：A prediction method based on LogSparse Transformer time series model is proposed to address the rapid changes and complexity of online public opinion in universities,in order to improve the accuracy and efficiency of public opinion management.The study used data from the Sina Weibo platform and constructed a LogSparse Transformer model through data preprocessing,correlation mining of long-term data,and long-range dependency modeling.The experimental results show that the LogSparse Transformer model outperforms traditional methods and machine learning algorithms in terms of prediction accuracy,while also having faster response speed and real-time processing capabilities.This model can effectively capture long-range dependencies in university public opinion events and reduce the time complexity of the model.This model provides a new and effective tool for predicting and managing public opinion in universities,which helps university managers to respond and manage online public opinion in a timely manner.
Keywords：online public opinionLogSparse Transformer time series modeluniversity managementdata cleaningattention mechanism
Publication Date：2025-05-31
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：8( 61-68 )