ECQED: Emotion-Cause Quadruple Extraction in Dialogs
Preprint
Abstract
The existing emotion-cause pair extraction (ECPE) task, unfortunately, ignores extracting the emotion type and cause type, while these fine-grained meta-information can be practically useful in real-world applications, i.e., chat robots and empathic dialog generation. Also the current ECPE is limited to the scenario of single text piece, while neglecting the studies at dialog level that should have more realistic values. In this paper, we extend the ECPE task with a broader definition and scenario, presenting a new task, Emotion-Cause Quadruple Extraction in Dialogs (ECQED), which requires detecting emotion-cause utterance pairs and emotion and cause types. We present an ECQED model based on a structural and semantic heterogeneous graph as well as a parallel grid tagging scheme, which advances in effectively incorporating the dialog context structure, meanwhile solving the challenging overlapped quadruple issue. Via experiments we show that introducing the fine-grained emotion and cause features evidently helps better dialog generation. Also our proposed ECQED system shows exceptional superiority over baselines on both the emotion-cause quadruple or pair extraction tasks, meanwhile being highly efficient.Keywords
All Related Versions
This publication has 32 references indexed in Scilit:
- Multimodal Emotion-Cause Pair Extraction in ConversationsIEEE Transactions on Affective Computing, 2022
- EmoInt-Trans: A Multimodal Transformer for Identifying Emotions and Intents in Social ConversationsIEEE/ACM Transactions on Audio, Speech, and Language Processing, 2022
- Public opinion monitoring through collective semantic analysis of tweetsSocial Network Analysis and Mining, 2022
- Recurrent synchronization network for emotion-cause pair extractionKnowledge-Based Systems, 2022
- PsychBERT: A Mental Health Language Model for Social Media Mental Health Behavioral Analysis2021
- Recognizing Emotion Cause in ConversationsCognitive Computation, 2021
- A Unified Target-Oriented Sequence-to-Sequence Model for Emotion-Cause Pair ExtractionIEEE/ACM Transactions on Audio, Speech, and Language Processing, 2021
- Speech Emotion Recognition Considering Nonverbal Vocalization in Affective ConversationsIEEE/ACM Transactions on Audio, Speech, and Language Processing, 2021
- CTNet: Conversational Transformer Network for Emotion RecognitionIEEE/ACM Transactions on Audio, Speech, and Language Processing, 2021
- Multiple Level Hierarchical Network-Based Clause Selection for Emotion Cause ExtractionIEEE Access, 2019