Research on the evolution of college online public opinion risk based on improved Grey Wolf Optimizer combined with LSTM model

Chao Cao, Ziyu Li, Lingzhi Li and Fanglu Luo

PLOS ONE, 2025, vol. 20, issue 1, 1-20

Abstract: Since the dissemination of information is more rapid and the scale of users on online platforms is enormous, the public opinion risk is more visible and harder to tackle for universities and authorities. Improving the accuracy of predictions regarding online public opinion crises, especially those related to campuses, is crucial for maintaining social stability. This research proposes a public opinion crisis prediction model that applies the Grey Wolf Optimizer (GWO) algorithm combined with long short-term memory (LSTM) and implements it to analyze a trending topic on Sina Weibo to validate its prediction accuracy. A full-chain analytical framework for online public opinion prediction is established in this study, which enables the model to illustrate the level of risk related to public opinion and its variation trend by introducing the public opinion risk index. The prediction accuracy of the model is validated through several evaluation criteria, and a comparison between real and predicted results, and the simulation of the intervention on this incident indicates that the proposed model is competent for both trend prediction and assisting in intervention. The study also demonstrates the importance of immediate response and intervention to public opinion crises.

Date: 2025
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0311749 (text/html)
https://journals.plos.org/plosone/article/file?id= ... 11749&type=printable (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:plo:pone00:0311749

DOI: 10.1371/journal.pone.0311749

Access Statistics for this article

More articles in PLOS ONE from Public Library of Science
Bibliographic data for series maintained by plosone ().