Hidden Emotions: Country-Based Fuzzy Logic Emotion Detection Method on Twitter Text
Wanyi Li, Yu Liu, Keqing Deng, Xiaokun Wu · 2025
Social networks provide valuable insights into various domains, including political dynamics, social issues, and business trends. Emotion detection on platforms like Twitter has become an effective method for gauging public sentiment. However, much of the existing literature predominantly focuses on primary emotions, often neglecting the subtle but significant influence of secondary emotions on event perception and public discourse. This study aims to address this gap by analyzing both primary and secondary emotional responses to the “Fukushima wastewater discharge” incident in 2023. A total of 76,285 tweets and comments related to the incident were collected from Twitter, offering a comprehensive dataset for sentiment analysis. Employing a fuzzy logic-based emotion detection framework, the study identifies secondary emotions embedded within the text and examines their role in shaping public emotional responses. The methodology includes several stages: data preprocessing to remove noise, feature extraction, sentiment analysis, and the application of fuzzy logic to detect secondary emotions, thereby enhancing traditional emotion classification techniques and providing a deeper understanding of public sentiment. The results underscore the significant role of secondary emotions in shaping public opinion and offer valuable implications for public opinion management, government decision-making, national image building, and advancements in natural language processing (NLP). This research introduces a novel approach to emotion detection, contributing to a more nuanced understanding of emotional dynamics in public discourse.