A multi-view network for real-time emotion recognition in conversations

作者：

Highlights：

• MVN is proposed to capture the word- and utterance-level dependencies for RTERC.

• MVN explores the emotion representation of a query from two different views.

• The experiments show that MVN outperforms the state-of-the-art baselines.

摘要

•MVN is proposed to capture the word- and utterance-level dependencies for RTERC.•MVN explores the emotion representation of a query from two different views.•The experiments show that MVN outperforms the state-of-the-art baselines.

论文关键词：Emotion recognition,Real-time conversations,Multi-view learning,Word-level dependencies,Utterance-level dependencies

论文评审过程：Received 26 May 2021, Revised 11 November 2021, Accepted 12 November 2021, Available online 27 November 2021, Version of Record 2 December 2021.

论文官网地址：https://doi.org/10.1016/j.knosys.2021.107751