| ISSN: 2583-3812 | ISSN: 2583-3812JOURNAL ARTICLE
Keywords: Great Recall, PTT, Emotional Polarization, Computational Social Science, Topic Modeling, Sentiment Analysis, entity Volume analysis.
Abstract: On July 26, 2025, a huge recall movement led by citizen organizations claiming to speak for the public was ultimately defeated by a devastating 25:0 margin. In addition to seriously harming the ruling party that worked to advance it, this case posed a serious challenge to the widely held political belief that "citizen groups represent public opinion." What causes a significant disparity between the actual vote results and a seemingly seamless political mobilization in the media and ground warfare? This study focuses on the PTT forum, a significant political and public opinion gathering spot in Taiwan, in order to address this central topic. It gathers important data from the two weeks prior to and following the recall vote in order to perform a thorough public opinion analysis. We investigate at three levels using techniques including topic modeling, sentiment analysis, and entity Volume analysis, and: (1) Narrative core: analyzing public opinion's discussion focus on the recall; (2) Emotional Flow: describing a community's collective emotional curve during the course of an event; and (3) Political shock: assessing the event's predicted influence on Taiwan's political landscape going forward. The purpose of this study is to demonstrate that online public opinion operations or traditional political mobilization may not be sufficient to fully penetrate the complex structure of public opinion in today's quickly changing information environment. The key to comprehending the outcome of this recall case and possibly even getting insight into the political climate of modern-day Taiwan region may lie in the underappreciated wisdom of the public, democratic literacy, intense social emotions, and the possible spiral effect of silence.
Article Info: Received: 11 Oct 2025; Received in revised form: 08 Nov 2025; Accepted: 11 Nov 2025; Available online: 16 Nov 2025
DOI: 10.22161/ijeel.4.6.7
| Total View: 233 | Downloads: 1 | Page No: 55-67 |  |
