Elamur Research⌕
☰
home / papers / 2021 / Directed Acyclic Graph Network for Conversational Emotion Recognition

Directed Acyclic Graph Network for Conversational Emotion Recognition

Summary

The paper tackles emotion recognition in conversation (ERC), where each utterance in a dialogue gets an emotion label. Existing approaches fall into two groups. Graph-based models look at a fixed window of nearby utterances and lose distant context and ordering. Recurrence-based models track history over time, but each step draws on only a few recent turns. The authors combine both strengths by treating a dialogue as a directed acyclic graph (DAG) in which information moves only forward in time.

Edges are built using speaker identity and position. Each utterance links back to every utterance since the ω-th most recent turn by the same speaker. That earlier turn is assumed to carry the older background context. On this graph they build DAG-ERC, an extension of DAGNN with two additions: a speaker-aware relation transformation and a contextual information unit. Utterance features come from a pre-trained Transformer language model.

On four benchmarks, DAG-ERC matches or beats strong baselines. It also stays stable as layers are added, unlike standard GNNs that suffer from over-smoothing.

Key points

  • Evaluated on IEMOCAP, MELD, DailyDialog and EmoryNLP. In the ablation table, the full model reaches F1 of 59.33 on DailyDialog and 39.02 on EmoryNLP.
  • Edges have two relation types: same speaker and different speaker. The ablation suggests this split is enough for two-party dialogues but too coarse for multi-party ones (MELD, EmoryNLP).
  • With more layers on IEMOCAP, RGAT-RoBERTa degrades sharply beyond 6 layers. DAGNN and DAG-ERC fluctuate only within a narrow range.
  • Emotional shift remains hard. On IEMOCAP, accuracy is 57.98% on shifted samples versus 74.25% on non-shifted ones. The shifted-case result still beats DialogueRNN (52.5%) and DialogXL (55%).
  • Common errors include confusing similar emotions (e.g., frustrated vs anger) and over-predicting neutral. Code is released at github.com/shenwzh3/DAG-ERC.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

Fields: Natural language processing

Cite

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2021.acl-long.123

@inproceedings{shen2021directed,
  title = {{Directed Acyclic Graph Network for Conversational Emotion Recognition}},
  author = {Shen, Weizhou and Wu, Siyue and Yang, Yunyi and Quan, Xiaojun},
  booktitle = {Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)},
  year = {2021},
  doi = {10.18653/v1/2021.acl-long.123},
  url = {https://doi.org/10.18653/v1/2021.acl-long.123}
}

References

30 references, 9 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Devamanyu Hazarika, Soujanya Poria, Amir Zadeh, Erik Cambria, Louis-Philippe Morency, and Roger Zimmermann. 2018b. Conversational memory network for emotion recognition in dyadic dialogue videos. In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational… find our card
  2. Miltiadis Allamanis, Earl T Barr, Premkumar Devanbu, and Charles Sutton. 2018. A survey of machine learning for big code and naturalness. ACM Computing Surveys (CSUR), 51(4):1–37. find
  3. Carlos Busso, Murtaza Bulut, Chi-Chun Lee, Abe Kazemzadeh, Emily Mower, Samuel Kim, Jeannette N Chang, Sungbok Lee, and Shrikanth S Narayanan. 2008. Iemocap: Interactive emotional dyadic motion capture database. Language resources and evaluation, 42(4):335. find
  4. Ankush Chatterjee, Kedhar Nath Narahari, Meghana Joshi, and Puneet Agrawal. 2019. Semeval-2019 task 3: Emocontext contextual emotion detection in text. In Proceedings of the 13th International Workshop on Semantic Evaluation, pages 39–48. find our card
  5. Sheng-Yeh Chen, Chao-Chun Hsu, Chuan-Chun Kuo, Ting-Hao, Huang, and Lun-Wei Ku. 2018. Emotionlines: An emotion corpus of multi-party conversations. In 11th International Conference on Language Resources and Evaluation, LREC 2018, pages 1597–1601. find
  6. Maxwell Crouse, Ibrahim Abdelaziz, Cristina Cornelio, Veronika Thost, Lingfei Wu, Kenneth Forbus, and Achille Fokoue. 2019. Improving graph neural network representations of logical formulae with subgraph pooling. arXiv preprint arXiv:1911.06904. — [arXiv:1911.06904](https://arxiv.org/abs/1911.06904) find
  7. Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina N. Toutanova. 2018. Bert: Pre-training of deep bidirectional transformers for language understanding. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologi… find
  8. Taichi Ishiwatari, Yuki Yasuda, Taro Miyazaki, and Jun Goto. 2020. Relation-aware graph attention networks with relational position encodings for emotion recognition in conversations. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 7360– 7370. find
  9. Wenxiang Jiao, Haiqin Yang, Irwin King, and Michael R Lyu. 2019. Higru: Hierarchical gated recurrent units for utterance-level emotion recognition. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volum… find
  10. Thomas N Kipf and Max Welling. 2016. Semi-supervised classification with graph convolutional networks. arXiv preprint arXiv:1609.02907. find our card
  11. Yanran Li, Hui Su, Xiaoyu Shen, Wenjie Li, Ziqiang Cao, and Shuzi Niu. 2017. Dailydialog: A manually labelled multi-turn dialogue dataset. In Proceedings of the Eighth International Joint Conference on Natural Language Processing (Volume 1: Long Papers), volume 1, pages 986–995. find our card
  12. Yinhan Liu, Myle Ott, Naman Goyal, Jingfei Du, Mandar Joshi, Danqi Chen, Omer Levy, Mike Lewis, Luke Zettlemoyer, and Veselin Stoyanov. 2019. Roberta: A robustly optimized bert pretraining approach. arXiv preprint arXiv:1907.11692. find
  13. Navonil Majumder, Pengfei Hong, Shanshan Peng, Jiankun Lu, Deepanway Ghosal, Alexander Gelbukh, Rada Mihalcea, and Soujanya Poria. 2020. Mime: Mimicking emotions for empathetic response generation. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pag… find our card
  14. Navonil Majumder, Soujanya Poria, Devamanyu Hazarika, Rada Mihalcea, Alexander Gelbukh, and Erik Cambria. 2019. Dialoguernn: An attentive rnn for emotion detection in conversations. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 33, pages 6818–6825. find
  15. Soujanya Poria, Devamanyu Hazarika, Navonil Majumder, Rada Mihalcea, Gautam Naik, and Erik Cambria. 2019b. Meld: A multimodal multi-party dataset for emotion recognition in conversations. In ACL 2019 : The 57th Annual Meeting of the Association for Computational Linguistics, pages 527–536. find our card
  16. Soujanya Poria, Navonil Majumder, Rada Mihalcea, and Eduard Hovy. 2019a. Emotion recognition in conversation: Research challenges, datasets, and recent advances. IEEE Access, 7:100943–100953. find
  17. Michael Sejr Schlichtkrull, Thomas N. Kipf, Peter Bloem, Rianne van den Berg, Ivan Titov, and Max Welling. 2018. Modeling relational data with graph convolutional networks. In 15th International Conference on Extended Semantic Web Conference, ESWC 2018, pages 593–607. find
  18. Björn Schuller, Michel Valster, Florian Eyben, Roddy Cowie, and Maja Pantic. 2012. find
  19. Avec 2012: the continuous audio/visual emotion challenge. In Proceedings of the 14th ACM international conference on Multimodal interaction, pages 449–456. find
  20. Weizhou Shen, Junqing Chen, Xiaojun Quan, and Zhixian Xie. 2020. Dialogxl: All-in-one xlnet for multi-party conversation emotion recognition. arXiv preprint arXiv:2012.08695. — [arXiv:2012.08695](https://arxiv.org/abs/2012.08695) find
  21. Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N. Gomez, Lukasz Kaiser, and Illia Polosukhin. 2017. Attention is all you need. In Proceedings of the 31st International Conference on Neural Information Processing Systems, pages 5998–6008. find
  22. Petar Veličković, Guillem Cucurull, Arantxa Casanova, Adriana Romero, Pietro Lio, and Yoshua Bengio. 2017. Graph attention networks. arXiv preprint arXiv:1710.10903. find our card
  23. Zhilin Yang, Zihang Dai, Yiming Yang, Jaime Carbonell, Russ R Salakhutdinov, and Quoc V Le. 2019. Xlnet: Generalized autoregressive pretraining for language understanding. In Advances in Neural Information Processing Systems, pages 5753–5763. find our card
  24. Sayyed M. Zahiri and Jinho D. Choi. 2017. Emotion detection on tv show transcripts with sequence-based convolutional neural networks. In AAAI Workshops, pages 44–52. find
  25. Dong Zhang, Liangqing Wu, Changlong Sun, Shoushan Li, Qiaoming Zhu, and Guodong Zhou. 2019a. Modeling both context- and speaker-sensitive dependence for emotion detection in multi-speaker conversations. In Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence, IJ… find
  26. Muhan Zhang, Shali Jiang, Zhicheng Cui, Roman Garnett, and Yixin Chen. 2019b. D-vae: A variational autoencoder for directed acyclic graphs. In Advances in Neural Information Processing Systems, pages 1588–1600. find
  27. Peixiang Zhong, Di Wang, and Chunyan Miao. 2019. Knowledge-enriched transformer for emotion detection in textual conversations. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-I… find our card
  28. Bing Shuai, Zhen Zuo, Bing Wang, and Gang Wang. 2016. Dag-recurrent neural networks for scene labeling. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 3620–3629. find
  29. Kai Sheng Tai, Richard Socher, and Christopher D Manning. 2015. Improved semantic representations from tree-structured long short-term memory networks. In Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Lan… find
  30. Veronika Thost and Jie Chen. 2021. Directed acyclic graph neural networks. In International Conference on Learning Representations. find

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/shen-2021-directed-acyclic-graph-network-conversational/