Read the full-text of this article for free

Title: Media perception and emotional evolution of Weibo disaster events based on TextCNN

Authors: Aimin Xie; Jianqiang Dai

Addresses: School of Culture and Media, Jiangxi Institute of Technology, Nanchang, 330098, Jiangxi, China ' School of Education, Jiangxi Institute of Technology, Nanchang, 330098, Jiangxi, China

Abstract: In the information age, Weibo is a key platform for catastrophic event information sharing and public emotional expression. This study takes the 2022 China Eastern Airlines passenger plane accident as a case study, based on the theory framework of media perception, and uses the TextCNN model to conduct sentiment analysis and emotional evolution research on accident related Weibo data. The results show that the accuracy of the model test set is 85.87%, the recall rate is 84.87%, the precision rate is 84.12%, and the F1 value is 84.49%, which is better than BiLSTM and CNN-BiLSTM. The overall proportion of negative emotions is 39.27%, positive is 38.11%, and neutral is 22.61%. Research reveals that information supply, social interaction, and cultural context drive emotional evolution, which can provide support for phased guidance and group governance of disaster public opinion.

Keywords: TextCNN; catastrophic events; Weibo; media perception.

DOI: 10.1504/IJRIS.2026.155469

International Journal of Reasoning-based Intelligent Systems, 2026 Vol.18 No.18, pp.29 - 40

Received: 31 Dec 2025
Accepted: 29 Apr 2026
Published online: 03 Aug 2026 *