DHF-Net: A hierarchical feature interactive fusion network for dialogue emotion recognition

作者：

Highlights：

• A novel model for emotion recognition from dialogue is introduced.

• The integration of context- and fine-grained information improves performance.

• A hybrid neural network group that connects with gate recurrent unit is proposed.

摘要

•A novel model for emotion recognition from dialogue is introduced.•The integration of context- and fine-grained information improves performance.•A hybrid neural network group that connects with gate recurrent unit is proposed.

论文关键词：Dialogue emotion recognition,Contextual information,Fine-grained information,Hierarchical feature,Interactive fusion

论文评审过程：Received 7 December 2021, Revised 31 July 2022, Accepted 10 August 2022, Available online 16 August 2022, Version of Record 24 August 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.118525