Emotion Arc

Emotion arc research focuses on modeling and predicting the dynamic changes in emotional states over time within various contexts, such as narratives, conversations, and social media interactions. Current research employs deep learning architectures, including transformers and graph convolutional networks, to analyze textual and visual data, often incorporating techniques like weakly supervised learning to improve prediction accuracy of continuous emotion dimensions (valence and arousal). This work has significant implications for improving human-computer interaction, enhancing sentiment analysis across multiple languages, and offering new insights into the emotional dynamics of communication and storytelling.