Knowledge-Interactive Network with Sentiment Polarity Intensity-Aware Multi-Task Learning for Emotion Recognition in Conversations

Yunhe Xie, Kailai Yang, Chengjie Sun, Bingquan Liu, Zhenzhou Ji


Abstract
Emotion Recognition in Conversation (ERC) has gained much attention from the NLP community recently. Some models concentrate on leveraging commonsense knowledge or multi-task learning to help complicated emotional reasoning. However, these models neglect direct utterance-knowledge interaction. In addition, these models utilize emotion-indirect auxiliary tasks, which provide limited affective information for the ERC task. To address the above issues, we propose a Knowledge-Interactive Network with sentiment polarity intensity-aware multi-task learning, namely KI-Net, which leverages both commonsense knowledge and sentiment lexicon to augment semantic information. Specifically, we use a self-matching module for internal utterance-knowledge interaction. Considering correlations with the ERC task, a phrase-level Sentiment Polarity Intensity Prediction (SPIP) task is devised as an auxiliary task. Experiments show that all knowledge integration, self-matching and SPIP modules improve the model performance respectively on three datasets. Moreover, our KI-Net model shows 1.04% performance improvement over the state-of-the-art model on the IEMOCAP dataset.
Anthology ID:
2021.findings-emnlp.245
Volume:
Findings of the Association for Computational Linguistics: EMNLP 2021
Month:
November
Year:
2021
Address:
Punta Cana, Dominican Republic
Editors:
Marie-Francine Moens, Xuanjing Huang, Lucia Specia, Scott Wen-tau Yih
Venue:
Findings
SIG:
SIGDAT
Publisher:
Association for Computational Linguistics
Note:
Pages:
2879–2889
Language:
URL:
https://aclanthology.org/2021.findings-emnlp.245
DOI:
10.18653/v1/2021.findings-emnlp.245
Bibkey:
Cite (ACL):
Yunhe Xie, Kailai Yang, Chengjie Sun, Bingquan Liu, and Zhenzhou Ji. 2021. Knowledge-Interactive Network with Sentiment Polarity Intensity-Aware Multi-Task Learning for Emotion Recognition in Conversations. In Findings of the Association for Computational Linguistics: EMNLP 2021, pages 2879–2889, Punta Cana, Dominican Republic. Association for Computational Linguistics.
Cite (Informal):
Knowledge-Interactive Network with Sentiment Polarity Intensity-Aware Multi-Task Learning for Emotion Recognition in Conversations (Xie et al., Findings 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.findings-emnlp.245.pdf
Video:
 https://aclanthology.org/2021.findings-emnlp.245.mp4
Data
ConceptNetDailyDialogIEMOCAPMELD