Learning contextual information and task alignment for emotion cause extraction in conversation

Abstract

Recognizing the fine-grained emotion cause extraction-Causal Span Extraction (CSE)-is challenging due to the more detailed analysis of contextual information. The previous research on CSE has focused on the powerful semantic representation capability of PLMs while overlooking the semantic coherence of the speaker and content when emotions arise in conversation. In this paper, we introduce a novel method by learning contextual information and enhancing the consistency of cross-task alignment. Specifically, we integrate the coreference resolution into the attention mechanism to capture the coreference-aware semantic correlations and employ the position relation strategy at both the utterance and token levels to understand the contextual information. Furthermore, by incorporating auxiliary tasks and a novel cross-task alignment approach, we reduce inconsistent predictions across tasks, thereby enabling a comprehensive, multi-dimensional comprehension of conversations. Our method demonstrates a marked improvement over current state-of-the-art models, evidenced by superior performance on two benchmark datasets.

Graphical abstract

Keywords

emotion cause / token level / spoken dialogues

References

[1] Young T, Xing F, Pandelea V, Ni J, Cambria E. Fusing task-oriented and open-domain dialogues in conversational agents. In: Proceedings of the 36th AAAI Conference on Artificial Intelligence. 2022, 11622–11629

[2] Tu G, Wen J, Liu C, Jiang D, Cambria E . Context-and sentiment-aware networks for emotion recognition in conversation. IEEE Transactions on Artificial Intelligence, 2022, 3( 5): 699–708

[3] Tu G, Liang B, Qin B, Wong K F, Xu R. An empirical study on multiple knowledge from ChatGPT for emotion recognition in conversations. In: Proceedings of Findings of the Association for Computational Linguistics: EMNLP 2023. 2023, 12160–12173

[4] Kim H, Kim B, Kim G. Perspective-taking and pragmatics for generating empathetic responses focused on emotion causes. In: Proceedings of 2021 Conference on Empirical Methods in Natural Language Processing. 2021, 2227–2240

[5] Lee S Y M, Chen Y, Huang C R. A text-driven rule-based system for emotion cause detection. In: Proceedings of the NAACL HLT 2010 Workshop on Computational Approaches to Analysis and Generation of Emotion in Text. 2010, 45–53

[6] Diao Y, Lin H, Yang L, Fan X, Chu Y, Wu D, Xu K . Emotion cause detection with enhanced-representation attention convolutional-context network. Soft Computing, 2021, 25( 2): 1297–1307

[7] Ping H, Zhang D, Zhu S, Li J, Zhou G . A benchmark for hierarchical emotion cause extraction in spoken dialogues. IEEE Signal Processing Letters, 2023, 30: 558–562

[8] Bhat A, Modi A. Multi-task learning framework for extracting emotion cause span and entailment in conversations. In: Proceedings of the 1st Transfer Learning for Natural Language Processing Workshop. 2023, 33–51

[9] Poria S, Majumder N, Hazarika D, Ghosal D, Bhardwaj R, Jian S Y B, Hong P, Ghosh R, Roy A, Chhaya N, Gelbukh A, Mihalcea R . Recognizing emotion cause in conversations. Cognitive Computation, 2021, 13( 5): 1317–1332

[10] Li J, Meng F, Lin Z, Liu R, Fu P, Cao Y, Wang W, Zhou J. Neutral utterances are also causes: enhancing conversational causal emotion entailment with social commonsense knowledge. In: Proceedings of the 31st International Joint Conference on Artificial Intelligence. 2022, 4209–4215

[11] Zhao W, Zhao Y, Li Z, Qin B. Knowledge-bridged causal interaction network for causal emotion entailment. In: Proceedings of the 37th AAAI Conference on Artificial Intelligence. 2023, 14020–14028

[12] Kong D, Yu N, Yuan Y, Fu G, Gong C. Discourse-aware emotion cause extraction in conversations. 2022, arXiv preprint arXiv: 2210.14419

[13] Gu X, Lou R, Sun L, Li S. PAGE: a position-aware graph-based model for emotion cause entailment in conversation. In: Proceedings of 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). 2023, 1–5

[14] Jiang D, Liu H, Tu G, Wei R . Window transformer for dialogue document: a joint framework for causal emotion entailment. International Journal of Machine Learning and Cybernetics, 2023, 14( 8): 2697–2707

[15] Zhang D, Yang Z, Meng F, Chen X, Zhou J. TSAM: a two-stream attention model for causal emotion entailment. In: Proceedings of the 29th International Conference on Computational Linguistics. 2022, 6762–6772

[16] Kullback S, Leibler R A . On information and sufficiency. The Annals of Mathematical Statistics, 1951, 22( 1): 79–86

[17] Zhang D, Wu L, Sun C, Li S, Zhu Q, Zhou G. Modeling both context-and speaker-sensitive dependence for emotion detection in multi-speaker conversations. In: Proceedings of the 28th International Joint Conference on Artificial Intelligence. 2019, 5415–5421

[18] Ishiwatari T, Yasuda Y, Miyazaki T, Goto J. Relation-aware graph attention networks with relational position encodings for emotion recognition in conversations. In: Proceedings of 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP). 2020, 7360–7370

[19] Shen W, Wu S, Yang Y, Quan X. Directed acyclic graph network for conversational emotion recognition. In: 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). 2021, 1551–1560

[20] Bansal K, Agarwal H, Joshi A, Modi A. Shapes of emotions: Multimodal emotion recognition in conversations via emotion shifts. In: Proceedings of the First Workshop on Performance and Interpretability Evaluations of Multimodal, Multipurpose, Massive-Scale Models. 2022, 44–56

[21] Fu Y, Yuan S, Zhang C, Cao J . Emotion recognition in conversations: a survey focusing on context, speaker dependencies, and fusion methods. Electronics, 2023, 12( 22): 4714

[22] Zheng X, Liu Z, Zhang Z, Wang Z, Wang J. UECA-prompt: universal prompt for emotion cause analysis. In: Proceedings of the 29th International Conference on Computational Linguistics. 2022, 7031–7041

[23] Wu Z, Dai X, Xia R . Pairwise tagging framework for end-to-end emotion-cause pair extraction. Frontiers of Computer Science, 2023, 17( 2): 172314

[24] Su X, Huang Z, Zhao Y, Chen Y, Dou Y, Pan H . Recent trends in deep learning based textual emotion cause extraction. IEEE/ACM Transactions on Audio, Speech, and Language Processing, 2023, 31: 2765–2786

[25] Li X, Gao W, Feng S, Zhang Y, Wang D. Boundary detection with BERT for span-level emotion cause analysis. In: Proceedings of Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021. 2021, 676–682

[26] Li X, Gao W, Feng S, Wang D, Joty S. Span-level emotion cause analysis by BERT-based graph attention network. In: Proceedings of the 30th ACM International Conference on Information & Knowledge Management. 2021, 3221–3226

[27] Li X, Gao W, Feng S, Wang D, Joty S. Span-level emotion cause analysis with neural sequence tagging. In: Proceedings of the 30th ACM International Conference on Information & Knowledge Management. 2021, 3227–3231

[28] Liu Y, Ott M, Goyal N, Du J, Joshi M, Chen D, Levy O, Lewis M, Zettlemoyer L, Stoyanov V. Roberta: a robustly optimized BERT pretraining approach. 2019, arXiv preprint arXiv: 1907.11692

[29] Toshniwal S, Wiseman S, Ettinger A, Livescu K, Gimpel K. Learning to ignore: long document coreference with bounded memory neural networks. In: Proceedings of 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP). 2020, 8519–8526

[30] Li Y, Su H, Shen X, Li W, Cao Z, Niu S. DailyDialog: a manually labelled multi-turn dialogue dataset. In: Proceedings of the 8th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2017, 986–995

[31] Ding Z, Xia R, Yu J. ECPE-2D: emotion-cause pair extraction based on joint two-dimensional representation, interaction and prediction. In: Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics. 2020, 3161–3170

[32] Ding Z, Xia R, Yu J. End-to-end emotion-cause pair extraction based on sliding window multi-label learning. In: Proceedings of 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP). 2020, 3574–3583

[33] Wei P, Zhao J, Mao W. Effective inter-clause modeling for end-to-end emotion-cause pair extraction. In: Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics. 2020, 3171–3181

[34] Yan H, Gui L, Pergola G, He Y. Position bias mitigation: a knowledge-aware graph model for emotion cause extraction. In: 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). 2021, 3364–3375

[35] Turcan E, Wang S, Anubhai R, Bhattacharjee K, Al-Onaizan Y, Muresan S. Multi-task learning and adapted knowledge models for emotion-cause extraction. In: Proceedings of Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021. 2021, 3975–3989

[36] Li J, Lin Z, Fu P, Wang W. Past, present, and future: conversational emotion recognition through structural modeling of psychological knowledge. In: Proceedings of Findings of the Association for Computational Linguistics: EMNLP 2021. 2021, 1204–1214

[37] Yang K, Zhang T, Ji S, Ananiadou S. A bipartite graph is all we need for enhancing emotional reasoning with commonsense knowledge. In: Proceedings of the 32nd ACM International Conference on Information and Knowledge Management. 2023, 2917–2927

[38] Huang B, Zhang Z, Zhao H. Tracing origins: coreference-aware machine reading comprehension. In: Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2022, 1281–1292

RIGHTS & PERMISSIONS

Higher Education Press