Methodology for visual analysis of psychological tension in online discourse on coronavirus vaccination

Purpose This study aims to introduce a novel methodology for visually analyzing psychological tension in social networks, particularly in the context of disturbances related to coronavirus vaccination. It also aims to enhance the interpretation of online discourse dynamics by integrating mathematical, linguistic and visual-analytical methods. Design/methodology/approach The study uses a comprehensive approach, including tweet array generation via the Vicinitas API, key term extraction, sentiment analysis and visualization tools such as Word-Cloud, VosViewer, Gephisto and Gephi. This methodology is tested on protests against coronavirus vaccination to evaluate its effectiveness in capturing the intricacies of digital discourse. Findings The study identifies key themes, including psychological stress linked to vaccination, protest movements and sentiments regarding trust and belief. Hierarchical representations and visualizations reveal the nuances within digital discourse, demonstrating the methodology’s capacity to discern patterns in social media interactions. Practical implications Practically, this methodology offers a robust tool for monitoring and interpreting online discourse, particularly in scenarios demanding immediate response, such as public health crises. Public health officials can use this method to detect early signs of misinformation or psychological distress, enabling timely interventions. Policymakers and analysts can leverage these insights to design communication strategies that build trust and mitigate public anxiety, aligning with societal needs for transparency and accountability. Originality/value This research introduces a novel integration of computational tools and expert insights, specifically designed for real-time analysis of psychological tension in online discourse. Unlike previous methods that focus on text analysis or sentiment evaluation independently, this approach uniquely combines sentiment dynamics with predictive modeling, offering a comprehensive lens for understanding digital interactions during sensitive public health crises.