Title: An evolution trend evaluation of social media network public opinion based on unsupervised learning

Authors: Yanhua Shen

Addresses: School of Public Management, China University of Mining and Technology, Xuzhou, Jiangsu, 221116, China; Nanjing Forestry University, Nanjing, Jiangsu, 210037, China

Abstract: In order to overcome the problems of poor evaluation effect, low accuracy and time-consuming of traditional methods, an evolution trend evaluation method of social media network public opinion based on unsupervised learning is proposed. Firstly, we establish the evaluation index system of public opinion evolution trend of social media network. Then, the graph convolution neural network is used to combine the evaluation index with neighbourhood features to extract the mixed features of public opinion evolution trend, and the correlation degree of mixed features is calculated by correlation ranking method. Finally, the evaluation model of public opinion evolution trend based on unsupervised learning is constructed according to the correlation degree of mixed features, and the evaluation results are obtained. The experimental results show that the proposed method has good evaluation effect of public opinion evolution trend, high evaluation accuracy and short evaluation time.

Keywords: unsupervised learning; social media networks; evolution of network public opinion; evolution trend assessment.

DOI: 10.1504/IJWBC.2024.136653

International Journal of Web Based Communities, 2024 Vol.20 No.1/2, pp.139 - 152

Received: 23 Mar 2022
Accepted: 09 Jun 2022
Published online: 15 Feb 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article