Research on the Public Opinion Perception and Behaviors for Emergencies Based on AI-Driven Emotional Interaction ——A Case Study of Regarding the "2·8 Yunlian Landslide"

Lv Rongjian¹, Gao Qiyue^2*

1.School of Business Administration, University of Science and Technology Liaoning; 2.School of Computer Science and Software Engineering, University of Science and Technology Liaoning

Abstract：

This paper uses the "2·8 Junlian Landslide" as a case study and employs sentiment analysis methods to examine the evolution of public sentiment regarding the incident. The findings reveal that while positive emotions dominated online discourse surrounding the event, localized pockets of negative sentiment were present, resulting in emotional volatility and polarization that could easily lead to public misperceptions. Public safety governance must integrate physical response with online public opinion management, effectively carry out on-site emergency response and disaster prevention efforts, and leverage intelligent monitoring to conduct public opinion early warning, positive guidance, and targeted intervention. This study provides theoretical references and practical support for enhancing public safety governance capabilities in response to emergencies.

Key Words：

AI-driven emotional interaction; "2·8 Junlian Landslide"; Weibo public opinion; emergency