Risks prevention and control of university online public opinion based on data clustering algorithm

Abstract

The risk of online public opinion continues to exist after the epidemic. This paper proposes an information clustering method based on cloud computing to build a risk prevention and control model of online public opinion in colleges and universities. Firstly, a measurement model of network public opinion similarity is constructed, and a method to determine the optimal threshold of public opinion clustering is designed. Secondly, the public opinions on the same topic are clustered according to the similarity between opinions and the determined clustering threshold. Finally, using the parallel computing ability of cloud computing, public opinions on different topics can be gathered quickly and accurately to provide data support for the risk prevention and control of public opinions in universities. The experimental results show that this method can quickly obtain the public opinion on the network of colleges and universities, and has a high clustering accuracy. Therefore, this method can provide some support for the real-time monitoring of online public opinion in colleges and universities.