Search

The Mechanism of User Sentiment Changes on the Evolution of Online Public Opinion: A Case Study of the Fat Cat Incident

Author(s)

Jian Hu1, Yuanmin Feng2, Dan Chen1,*

Affiliation(s)

1School of Management, Chongqing University of Technology, Chongqing, China 2School of Language and Communication, Chongqing University of Technology, Chongqing, China *Corresponding Author

Abstract

Short-video platforms are now central to online public opinion, and user sentiment has become a key driver of opinion evolution. Using the “Fat Cat Incident” on TikTok as a case study, 59,915 comments were analysed with methods such as word segmentation and sentiment analysis. The study reveals a clear sentiment trajectory: polarisation → opposition → rationality → reflection. During the outbreak, anger and empathy drive agenda-setting and group polarisation. In the advancement stage, doubt and anger deepen opinion divides. At the reversal stage, an official announcement replaces emotional release with rational thinking. In the early decline stage, relief and reflection build rational consensus. In the late decline stage, deep reflection extends the discussion to broader social issues. User sentiment shapes public opinion through four pathways: agenda focusing, stance solidification, opinion reversal, and value extension. The target of sentiment shifts from condemning “Tan Zhu” to criticising chaotic online traffic manipulation. The study concludes that changes in user sentiment are a core force in short-video opinion evolution, offering insights for early warning, sentiment guidance, and rational management.

Keywords

Online Public Opinion; User Sentiment; Mechanism of Action; the Fat Cat Incident

References

[1] Tian J, Xiong H, Tang M, et al. Research on the Evolution of User Emotions and the Attribution of Negative Comments in Online Public Opinion Regarding University Emergencies. Journal of Intelligence Science, 2025, 43(1): 58-68. [2] Zhang T. Mechanisms of algorithmic recommendation on social media opinion polarization: An empirical analysis based on short video platforms. International Journal of e-Collaboration, 2026, 22(1): 1-16. [3] Wang L, Kim K. Analyzing group polarization through text emotion measurement and time series prediction: A comparative study across three online platforms. Measurement: Sensors, 2024, 33: 101216. [4] Xie X, Lin Y. A study on the influence of user comment format changes on public opinion formation. News and Writing, 2023, 00(03): 54. [5] Huang X, Zhao C. Application of text mining in online public opinion information analysis. Information Science, 2009, 27(1): 94-99. [6] Wang T, Yang W. Review of research on text sentiment analysis methods. Computer Engineering and Applications, 2021, 57(12): 11. [7] Ma D, Lu J, Zhu H. Research on social network users’ behaviour of disseminating others’ private information based on text emotion classification. Information Science, 2023, 41(2): 60. [8] Su F, Zhu H, Huang H. Research on the transmission mechanism of emotional polarisation under the hierarchical communication structure of social media——Based on web mining and empirical analysis of the “2·27 incident” on Weibo. Contemporary Communication, 2023(5): 99-103. [9] Xia D. “Anger Bait”: The affect mechanism in the algorithm age and emotional polarisation in the public sphere. Culture and Art Studies, 2026, 85(01): 55-63+114. [10] Xiao Z. Emotional resonance and symbol consumption: A study of symbol construction in variety shows——Taking “The Big Band” as an example. News and Writing, 2019, 0(9): 48-53. [11] Carpenter C J, Averbeck J M. What do superdiffusers do when they want to persuade someone about politics on facebook? Communication Quarterly, 2020, 68(1): 54-72.