Abstract: 【Purpose/significance】Constructing the LDA model of user reviews of university teachers’morals and public opinion on
Weibo.the identification of the evolution characteristics of public opinion and the analysis of the network topic propagation path under
this topic.can more accurately identify the evolution characteristics of public opinion and analyze the propagation path of key topics.
helping universities and education The department conducts public opinion supervision and public opinion guidance more effectively.
【Method/process】This article takes the event of "a professor of Tianjin University academic fraud" as an example.Based on the LDA
model.the topic generation model of Weibo users under the public opinion of teachers’ethics in colleges and universities is construct‐
ed.The confusion evaluation index is used to determine the optimal number of topics in the LDA model,and information entropy is used to determine each.The topic intensity of the topic on different dates is displayed through the keyword co-occurrence knowledge graph and word cloud to show the public opinion topic.and finally the topic transmission path is determined based on the topic similarity【. Re‐sult/conclusion】The microblog user comment topic generation model and co-occurrence knowledge graph based on the LDA model can analyze the important topic hotspots of the network user community.The LDA model and information entropy can accurately identi‐fy the evolutionary characteristics of public opinion.identify the optimal propagation path of the topic.and guide the public opinion【. In‐novation/limitation】The article innovatively constructs the LDA topic model of academic moral public opinion in colleges and universi‐ties.effectively determines the subject of Weibo user groups.identifies the evolutionary characteristics of public opinion.and analyzes the communication path between topics.It is universal.it further expands the analysis of other public opinions of college teachers’eth‐ics and the integrated network Public sentiment analysis is the next research content.