User-Level Emotion Detection Framework for Social Network Data

Social media interactions are inherently dynamic, with emotions fluctuating over time based on various events and user engagements. Traditional methods often overlook the temporal aspect of these interactions, limiting the depth of emotional analysis and the insights that can be derived. In this paper, we propose a novel temporal graph framework for detecting and analyzing emotions within social media networks. By leveraging the temporal nature of social media interactions, our framework incorporates both dynamic graph structures and sentiment analysis to track emotional trends over time. Proposed method demonstrate the effectiveness of our approach through empirical experiments on real-world social media data, revealing significant patterns and trends in user emotions. Our results highlight the potential of temporal graph models to enhance emotion detection and provide actionable insights for various applications, including sentiment-driven marketing and crisis management. The framework achieved an accuracy of 89% in detecting user emotions, a significant improvement over traditional static models, which averaged around 76%. Additionally, the temporal model identified emotional trends with a precision of 85% and a recall of 83%, compared to 70% precision and 68% recall achieved by baseline methods. These results indicate that the temporal graph framework not only tracks emotional trends more accurately over time but also provides deeper insights into the evolution of user sentiments within social networks.