Dialogue emotion recognition method based on dynamic graph network
Jun Hong Cao, Wei Xia, Simin Wang, Ziping Zhao · Proceedings of the 2021 5th International Conference on Electronic Information Technology and Computer Engineering · 2021
Emotion Recognition in Conversation has attracted widespread attention from researchers due to its wide application prospects, but most of the existing methods ignore that the dialogue structure will change as the dialogue progresses. To solve this problem, we propose a dialogue sentiment analysis method based on Dynamic Graph Neural Networks (DyGNN). We conduct experiments on two sentiment classification datasets thereby demonstrating that our model is superior to the state of the art in most cases.