Emotion recognition in conversations based on discourse parsing and graph attention network
The research on emotion recognition in conversations (ERC) focuses on the interrelationship between conversational context and speaker modeling. The current research usually ignores the dependency within the conversation, which leads to the weak connection between the context of the conversation and...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
Beijing Xintong Media Co., Ltd
2024-05-01
|
| Series: | Dianxin kexue |
| Subjects: | |
| Online Access: | http://www.telecomsci.com/zh/article/doi/10.11959/j.issn.1000-0801.2024149/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841528835918528512 |
|---|---|
| author | HAO Xiulan WEI Shaohua CAO Qian ZHANG Xiongtao |
| author_facet | HAO Xiulan WEI Shaohua CAO Qian ZHANG Xiongtao |
| author_sort | HAO Xiulan |
| collection | DOAJ |
| description | The research on emotion recognition in conversations (ERC) focuses on the interrelationship between conversational context and speaker modeling. The current research usually ignores the dependency within the conversation, which leads to the weak connection between the context of the conversation and the lack of logic between the speakers. Therefore, an emotion recognition in conversations model based on discourse parsing and graph attention network (DPGAT) was proposed to integrate the inter-dependency of conversation into the context modeling to make contextual information more dependent and global. Firstly, the discourse dependency relationships within the conversation were obtained through discourse parsing, and the discourse dependency graph and the speaker relationship graph were constructed. Subsequently, different types of speaker diagrams were internally integrated by multi-head attention mechanisms. Based on the graph attention network, cyclic learning was combined with dependency relationships to achieve the effective integration of contextual information and speaker information, realizing the external integration of context-related information in conversations. Finally, by analyzing the results of internal and external integration, the complete conversation context was restored, and the speaker's emotions were analyzed. By evaluating and verifying on English dataset MELD, EmoryNLP, DailyDialog and Chinese dataset M3ED, F1 scores were 66.23%, 40.03%, 59.28% and 52.77%, respectively. Compared with mainstream models, the proposed model at least reaches state-of-the-art, and can be used in different language scenarios. |
| format | Article |
| id | doaj-art-37c8611c276b4b319634dd761aa8a792 |
| institution | Kabale University |
| issn | 1000-0801 |
| language | zho |
| publishDate | 2024-05-01 |
| publisher | Beijing Xintong Media Co., Ltd |
| record_format | Article |
| series | Dianxin kexue |
| spelling | doaj-art-37c8611c276b4b319634dd761aa8a7922025-01-15T03:33:30ZzhoBeijing Xintong Media Co., LtdDianxin kexue1000-08012024-05-014010011160129683Emotion recognition in conversations based on discourse parsing and graph attention networkHAO XiulanWEI ShaohuaCAO QianZHANG XiongtaoThe research on emotion recognition in conversations (ERC) focuses on the interrelationship between conversational context and speaker modeling. The current research usually ignores the dependency within the conversation, which leads to the weak connection between the context of the conversation and the lack of logic between the speakers. Therefore, an emotion recognition in conversations model based on discourse parsing and graph attention network (DPGAT) was proposed to integrate the inter-dependency of conversation into the context modeling to make contextual information more dependent and global. Firstly, the discourse dependency relationships within the conversation were obtained through discourse parsing, and the discourse dependency graph and the speaker relationship graph were constructed. Subsequently, different types of speaker diagrams were internally integrated by multi-head attention mechanisms. Based on the graph attention network, cyclic learning was combined with dependency relationships to achieve the effective integration of contextual information and speaker information, realizing the external integration of context-related information in conversations. Finally, by analyzing the results of internal and external integration, the complete conversation context was restored, and the speaker's emotions were analyzed. By evaluating and verifying on English dataset MELD, EmoryNLP, DailyDialog and Chinese dataset M3ED, F1 scores were 66.23%, 40.03%, 59.28% and 52.77%, respectively. Compared with mainstream models, the proposed model at least reaches state-of-the-art, and can be used in different language scenarios.http://www.telecomsci.com/zh/article/doi/10.11959/j.issn.1000-0801.2024149/emotion recognition in conversationsdiscourse parsinggraph attention network |
| spellingShingle | HAO Xiulan WEI Shaohua CAO Qian ZHANG Xiongtao Emotion recognition in conversations based on discourse parsing and graph attention network Dianxin kexue emotion recognition in conversations discourse parsing graph attention network |
| title | Emotion recognition in conversations based on discourse parsing and graph attention network |
| title_full | Emotion recognition in conversations based on discourse parsing and graph attention network |
| title_fullStr | Emotion recognition in conversations based on discourse parsing and graph attention network |
| title_full_unstemmed | Emotion recognition in conversations based on discourse parsing and graph attention network |
| title_short | Emotion recognition in conversations based on discourse parsing and graph attention network |
| title_sort | emotion recognition in conversations based on discourse parsing and graph attention network |
| topic | emotion recognition in conversations discourse parsing graph attention network |
| url | http://www.telecomsci.com/zh/article/doi/10.11959/j.issn.1000-0801.2024149/ |
| work_keys_str_mv | AT haoxiulan emotionrecognitioninconversationsbasedondiscourseparsingandgraphattentionnetwork AT weishaohua emotionrecognitioninconversationsbasedondiscourseparsingandgraphattentionnetwork AT caoqian emotionrecognitioninconversationsbasedondiscourseparsingandgraphattentionnetwork AT zhangxiongtao emotionrecognitioninconversationsbasedondiscourseparsingandgraphattentionnetwork |