从认知学角度,隐喻情绪由句子中“源语义场景-目标语义场景”词对的情绪场景融合而成。鉴于此特点,该文提出了融合词语场景的隐喻情绪识别模型。该模型借助情绪词典及大规模语料库,构建了词语情绪分布表示获取算法,用于捕获句子中映射词对的情绪分布表示。在此基础上,利用注意力机制与最大池化策略,编码句子的多情绪场景融合表示,以刻画句子情绪形成的诱因。最后,设计情绪分类器,联合句子情绪及上下文表示作为输入,多角度地构建句子的语义,以提升隐喻情绪识别性能。在隐喻情绪数据集上进行实验,与基线模型和最好评测模型进行对比,该文提出的模型在宏F1值上提升了5.74%与2.73%。另外,定性的实例分析解释了词语场景对隐喻情绪识别的作用。
Abstract
According to cognitive science, metaphorical emotion is composed of the emotional scenes of the word pair "source semantic scene-target semantic scene" in a sentence. Therefore, this paper proposes a metaphorical emotion identification model by fusing word scenes. This method captures emotion distribution of word pairs with the help of emotion dictionary and large-scale corpus. Then, we use the attention and maxpool mechanisms to encode sentence representations integrated with multiple emotion scenes. Finally, an emotion classifier is designed by combining sentence emotion and context representation as input. Experimental results demonstrate the proposed model improves by 5.74% and 2.73% on macro-F value compared with baselines. In additon,qualitative case study explains the role of word scene in metaphorical emotion identification task.
关键词
词语场景 /
隐喻 /
情绪
{{custom_keyword}} /
Key words
word scene /
metaphor /
emotion
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] LAKOFF G, MARK J. Metaphors we live by[M]. University of Chicago Press, 2008.
[2] SHUTOVA E. Design and evaluation of metaphor processing systems[J]. Computational Linguistics, 2015, 41(4): 579-623.
[3] DANKERS V, REI M, LEWIS M, et al. Modelling the interplay of metaphor and emotion through multitask learning[C]//Proceedings of the Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing, 2019: 2218-2229.
[4] KANG L, LIU J, LIU L, et al. Semi-supervised emotion recognition in textual conversation via a context-augmented auxiliary training task[J]. Information Processing & Management, 2021, 58(6): 102717.
[5] LIN N, FU S, LIN X, et al. Multi-label emotion classification based on adversarial multi-task learning[J]. Information Processing & Management, 2022, 59(6): 103097.
[6] KLINGER R, DE CLERCQ O, MOHAMMAD S, et al. IEST:WASSA-2018 implicit emotions shared task[C]//Proceedings of the 9th Workshop on Computational Approaches to Subjectivity, Sentiment and Social Media Analysis, 2018: 31-42.
[7] CHEN X, HAI Z, LI D, et al. Jointly identifying rhetoric and implicit emotions via multi-task learning[C]//Proceedings of the Association for Computational Linguistics: ACL-IJCNLP, 2021: 1429-1434.
[8] LIAO J, WANG M, CHEN X, et al. Dynamic commonsense knowledge fused method for Chinese implicit sentiment analysis[J]. Information Processing & Management, 2022, 59(3): 102934.
[9] WANG Z, LI S, WU F, et al. Overview of NLPCC 2018 shared task 1: Emotion detection in code-switching text[C]//Proceedings of the CCF International Conference on Natural Language Processing and Chinese Computing. Springer, Cham, 2018: 429-433.
[10] YUE T, CHEN C, ZHANG S, et al. Ensemble of neural networks with sentiment words translation for code-switching emotion detection[C]//Proceedings of the CCF International Conference on Natural Language Processing and Chinese Computing. Springer, Cham, 2018: 411-419.
[11] AMEER I, BLC N, SIDDIQUI M H F, et al. Multi-label emotion classification in texts using transfer learning[J]. Expert Systems with Applications, 2023, 213: 118534.
[12] ZHANG D, WU L, SUN C, et al. Modeling both context-and speaker-sensitive dependence for emotion detection in multi-speaker conversations[C]//Proceedings of the IJCAI, 2019: 5415-5421.
[13] LU X, ZHAO Y, WU Y, et al. An iterative emotion interaction network for emotion recognition in conversations[C]//Proceedings of the 28th International Conference on Computational Linguistics, 2020: 4078-4088.
[14] KHANPOUR H, CARAGEA C. Fine-grained emotion detection in health-related online posts[C]//Proceedings of the Conference on Empirical Methods in Natural Language Processing, 2018: 1160-1166.
[15] DESAI S,CARAGEA C, Li J J. Detecting perceived emotions in hurricane disasters[C]//Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, 2020: 5290-5305.
[16] DELBROUCK J B, TITS N, BROUSMICHE M, et al. A transformer-based joint-encoding for emotion recognition and sentiment analysis[C]//Proceedings of the 2nd Grand-Challenge and Workshop on Multimodal Language, 2020: 1-7.
[17] ZHOU X, WANG Z, LI S, et al. Emotion detection with neural personal discrimination[C]//Proceedings of the Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing, 2019: 5499-5507.
[18] CHEN Y, HOU W, CHENG X, et al. Joint learning for emotion classification and emotion cause detection[C]//Proceedings of the Conference on Empirical Methods in Natural Language Processing, 2018: 646-651.
[19] RAJAMANICKAM S, MISHRA P,YANNAKOUDAKIS H, et al. Joint modelling of emotion and abusive language detection[C]//Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, 2020: 4270-4279.
[20] XIANG R, LU Q, JIAO Y, et al. Leveraging writing systems changes for deep learning based Chinese affective analysis[J]. International Journal of Machine Learning and Cybernetics, 2019, 10(11): 3313-3325.
[21] CHAUHAN D S, DHANUSH S R, EKBAL A, et al. Sentiment and emotion help sarcasm: A multi-task learning framework for multi-modal sarcasm, sentiment and emotion analysis[C]//Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, 2020: 4351-4360.
[22] UYMAZ H A, METIN S K. Vector based sentiment and emotion analysis from text: A survey[J]. Engineering Applications of Artificial Intelligence, 2022, 113: 104922.
[23] SARAVIA E, LIU H C T, HUANG Y H, et al. Carer: Contextualized affect representations for emotion recognition[C]//Proceedings of the Conference on Empirical Methods in Natural Language Processing, 2018: 3687-3697.
[24] GAONKAR R, KWON H,BASTAN M, et al. Modeling label semantics for predicting emotional reactions[C]//Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, 2020: 4687-4692.
[25] YING W, XIANG R, LU Q. Improving multi-label emotion classification by integrating both general and domain-specific knowledge[C]//Proceedings of the 5th Workshop on Noisy User-generated Text, 2019: 316-321.
[26] LI C, BAO Z, LI L, et al. Exploring temporal representations by leveraging attention-based bidirectional LSTM-RNNs for multi-modal emotion recognition[J]. Information Processing & Management, 2020, 57(3): 102185.
[27] ROZENTAL A, FLEISCHER D,KELRICH Z. Amobee at IEST 2018: Transfer learning from language models[C]//Proceedings of the 9th Workshop on Computational Approaches to Subjectivity, Sentiment and Social Media Analysis, 2018: 43-49.
[28] BALAZS J, MARRESE-TAYLOR E, MATSUO Y. IIIDYT at IEST 2018: Implicit emotion classification with deep contextualized word representations[C]//Proceedings of the 9th Workshop on Computational Approaches to Subjectivity, Sentiment and Social Media Analysis, 2018: 50-56.
[29] CHRONOPOULOU A,MARGATINA A, BAZIOTIS C, et al. NTUA-SLP at IEST 2018: Ensemble of neural transfer methods for implicit emotion classification[C]//Proceedings of the 9th Workshop on Computational Approaches to Subjectivity, Sentiment and Social Media Analysis, 2018: 57-64.
[30] LIAO J, WANG S, LI D. Identification of fact-implied implicit sentiment based on multi-level semantic fused representation[J]. Knowledge-Based Systems, 2019, 165: 197-207.
[31] WEI J, LIAO J, YANG Z, et al.BiLSTM with multi-polarity orthogonal attention for implicit sentiment analysis[J]. Neurocomputing,2020,383: 165-173.
[32] WU Y, ZHAO Y, LU X, et al. Modeling incongruity between modalities for multimodal sarcasm detection[J]. IEEE MultiMedia, 2021, 28(2): 86-95.
[33] CHEN X, HAI Z H, WANG S, et al. Metaphor identification: A contextual inconsistency based neural sequence labeling approach [J]. Neurocomputing, 2021, 428:268-279.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
基金
国家自然科学基金(62076158,62106130,62072294);山西省基础研究计划资助项目(202103021223267,20210302124084);山西省高等学校科技创新计划项目(2021L297,2021L284);太原科技大学科研启动基金(20212053,20222107);CCF-智谱AI大模型基金(CCF-zhipu202310)
{{custom_fund}}
PDF(3297 KB)
