利用异构图神经网络实现情绪-原因对的有效抽取

doi:10.11896/jsjkx.211100265

摘要/Abstract

摘要： 情绪-原因对的自动抽取,是文本情感分析的新任务,旨在以子句为单位,从不带任何标注的原始文本中识别情绪表达,并确定产生相应情绪的原因,形成情绪-原因对。完成上述任务的关键是有效捕捉情绪和原因之间以及不同情绪-原因对之间的关联。针对现有研究在捕捉这些关联方面存在的粒度过粗、无法有效区分不同子句对之间因果关系的相互影响等不足,提出了一种基于异构图神经网络的情绪-原因对抽取方法。该方法首先构建以子句和子句对为顶点的异构图,其中子句和子句对之间以及不同的子句对之间存在不同类型的边,用于捕捉各种细粒度的关联;然后采用带有注意力机制的异构图神经网络顶点表达更新算法,对子句和子句对的初始表达进行迭代更新;接着将更新后的子句对表达输入到二元分类器,通过该分类器判断相应的子句对是否存在情绪-原因关系。在情绪-原因对抽取任务的基准数据集上进行的实验表明,所提基于异构图神经网络的方法具有稳定的效果提升,在F1值上比当前最好的方法高0.85%;如果底层编码器(用于得到初始的子句表达和子句对表达)采用BERT,F1值可以达73.12%,也优于底层编码器同样采用BERT的现有最新算法。

关键词: 情感分析, 情绪原因对抽取, 异构图神经网络, 图神经网络

Abstract: As an emerging task in text sentiment analysis,the automatic extraction of emotion-cause pairs aims to identify emotion expression from the raw texts without any annotation in the unit of clauses,and identify the causes for the corresponding emotions to form emotion-cause pairs.The crucial point of this task is focused on how to effectively capture the relationship between emotions and causes and among different emotion-cause pairs.To overcome the shortcomings of existing researches in capturing these associations,such as too coarse granularity and unable to effectively distinguish the mutual influence of causal relations between different pairs,this paper proposes an emotion-cause pair extraction method based on a heterogeneous graph neural network.Initially,we construct a heterogeneous graph with clauses and clause pairs as vertices,in which there are different types of edges between clauses and clause pairs and between different clause pairs to capture various fine-grained associations.Then using the heterogeneous graph neural network algorithm with attention mechanism to iteratively update the vertex embeddings of clauses and clause pairs.Finally,the updated embeddings is input to the binary classifier,and the classifier judges whether the corresponding pair has an emotion-cause relationship.To evaluate the effectiveness of the proposed model,we conduct a series of experiments on a benchmark dataset of the emotion-cause pair extraction task.The results demonstrate that the method based on the heterogeneous graph neural network proposed in this paper has a stable effect improvement,and the F1 value is 0.85% higher than the state-of-art baselines.When the bottom encoder(for obtaining the initial embeddings of clauses and clause pairs) is replaced by BERT,the F1 value can reach 73.12%,and our model also outperforms the state-of-art algorithm.

Key words: Sentiment analysis, Emotion-Cause pair extraction, Heterogeneous graph neural network, Graph neural network

中图分类号:

TP391

蒲金垚, 卜令梅, 卢永美, 叶子铭, 陈黎, 于中华. 利用异构图神经网络实现情绪-原因对的有效抽取[J]. 计算机科学, 2023, 50(1): 205-212. https://doi.org/10.11896/jsjkx.211100265

PU Jinyao, BU Lingmei, LU Yongmei, YE Ziming, CHEN Li, YU Zhonghua. Utilizing Heterogeneous Graph Neural Network to Extract Emotion-Cause Pairs Effectively[J]. Computer Science, 2023, 50(1): 205-212. https://doi.org/10.11896/jsjkx.211100265

参考文献

[1]GUI L,WU D,XU R,et al.Event-driven emotion cause extraction with corpus construction[C]//Proceedings of the 2016 Conference on Empirical Methods in Natural Language Proces-sing.Texas:ACL,2016:1639-1649.
[2]XIA R,DING Z.Emotion-Cause Pair Extraction:A new task to emotion analysis in texts[C]//Proceedings of the 57th Confe-rence of the Association for Computational Linguistics.Flo-rence:ACL,2019:1003-1012.
[3]CHEN Y,HOU W,CHENG X,et al.Joint learning for emotion classification and emotion cause detection[C]//Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing.Brussels:ACL,2018:646-651.
[4]WEI P,ZHAO J,MAO W.Effective inter-clause modeling forEnd-to-End Emotion-Cause Pair Extraction[C]//Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics.Online:ACL,2020:3171-3181.
[5]SONG H,ZHANG C,LI Q,et al.End-to-end emotion-cause pair extraction via learning to link[J].arXiv:2002.10710,2020.
[6]DING Z,XIA R,YU J.End-to-End Emotion-Cause Pair Extraction based on sliding window Multi-Label Learning[C]//Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing(EMNLP).ACL,2020:3574-3583.
[7]WU S,CHEN F,WU F,et al.A Multi-Task Learning Neural Network for Emotion-Cause Pair Extraction[J/OL].ECAI.2020:2212-2219.https://ebooks.iospress.nl/doi/10.3233/FAIA200347.
[8]CHENG Z,JIANG Z,YIN Y,et al.A symmetric local searchnetwork for Emotion-Cause Pair Extraction[C]//Proceedings of the 28th International Conference on Computational Linguistics.Barcelona:ACL,2020:139-149.
[9]CHEN X,LI Q,WANG J.A Unified Sequence Labeling Model for Emotion Cause Pair Extraction[C]//Proceedings of the 28th International Conference on Computational Linguistics.Barcelona:ACL,2020:208-218.
[10]YUAN C,FAN C,BAO J,et al.Emotion-Cause Pair Extraction as sequence labeling based on a novel tagging scheme[C]//Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing(EMNLP).ACL,2020:3568-3573.
[11]FAN C,YUAN C,DU J,et al.Transition-based directed graph construction for emotion-cause pair extraction[C]//Proceedings of the 58th Annual Meeting of the Association for Computa-tional Linguistics.ACL,2020:3707-3717.
[12]DING Z,XIA R,YU J.ECPE-2D:Emotion-Cause Pair Extraction based on Joint Two-Dimensional Representation,Interaction and Prediction[C]//Proceedings of the 58th Annual Mee-ting of the Association for Computational Linguistics.ACL,2020:3161-3170.
[13]CHEN Y,HOU W,LI S,et al.End-to-End Emotion-Cause Pair Extraction with graph convolutional network[C]//Proceedings of the 28th International Conference on Computational Linguistics.Barcelona:ACL,2020:198-207.
[14]LEE S,CHEN Y,HUANG C.A text-driven rule-based system for emotion cause detection[C]//The NAACL HLT 2010 Workshop on Computational Approaches to Analysis and Ge-neration of Emotion in Text.California:ACL,2010:45-53.
[15]LI W,XU H.Text-based emotion classification using emotion cause extraction[J].Expert Systems with Applications,2014,41(4):1742-1749.
[16]GAO K,XU H,WANG J.A rule-based approach to emotion cause detection for Chinese micro-blogs[J].Expert Systems with Applications,2015,42(9):4517-4528.
[17]YADA S,IKEDA K,HOASHI K,et al.A bootstrap method for automatic rule acquisition on emotion cause extraction[C]//IEEE International Conference on Data Mining Workshops.New Orleans:IEEE,2017:414-421.
[18]GHAZI D,INKPEN D,SZPAKOWICZ S.Detecting emotionstimuli inemotion-bearing sentences[C]//Computational Linguistics and Intelligent Text Processing.Cairo:Spring,2015:152-165.
[19]SONG S,YAO M.Detecting concept-level emotion cause in microblogging[C]//Proceedings of the 24th International Confe-rence on World Wide Web Companion.New York:ACM,2015:119-120.
[20]CHEN Y,LEE S,LI S,et al.Emotion cause detection with linguistic constructions[C]//23rd International Conference on Computational Linguistics.Beijing:COLING,2010:179-187.
[21]GUI L,HU J,HE Y,et al.A question answering approach foremotion cause extraction[C]//Proceedings of the 2017 Confe-rence on Empirical Methods in Natural Language Processing.Copenhagen:ACL,2017:1593-1602.
[22]LI X,SONG K,FENG S,et al.A co-attention neural network model for emotion cause analysis with emotional context awareness[C]//Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing.Brussels:ACL,2018:4752-4757.
[23]XIA R,ZHANG M,DING Z.RTHN:A RNN-TransformerHierarchical Network for emotion cause extraction[C]//Procee-dings of the Twenty-Eighth International Joint Conference on Artificial Intelligence.Macao:IJCAI,2019:5285-5291.
[24]FAN C,YAN H,DU J,et al.A knowledge regularized hierarchical approach for emotion cause analysis[C]//Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing.Hong Kong:ACL,2019:5618-5628.
[25]DING Z,HE H,ZHANG M,et al.From independent prediction to re-ordered prediction:integrating relative position and global label information to emotion cause identification[C]//Procee-dings of the AAAI Conference on Artificial Intelligence.Ha-waii:AAAI,2019:6343-6350.
[26]CHEN Y,HOU W,CHENG X,et al.Joint learning for emotion classification and emotion cause detection[C]//Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing.Brussels:ACL,2018:646-651.
[27]QIU J,TANG J,MA H,et al.Deepinf:social influence prediction with deep learning[C]//Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining.New York:ACM,2018:2110-2119.
[28]SRIVASTAVA R K,GREFF K,SCHMIDHUBER J.Training very deepnetworks[J].arXiv:1507.06228,2015.
[29]TANG H,JI D,ZHOU Q.Joint multi-level attentional model for emotion detection and emotion-cause pairextraction[J].Neurocomputing,2020,409:329-340.
[30]DEVLIN J,CHANG M W,LEE K,et al.Bert:Pre-training of deep bidirectional transformers for language understanding[J].arXiv:1810.04805,2018.

相关文章 15

[1]	郝敬宇, 文静轩, 刘华锋, 景丽萍, 于剑. 结合全局信息的深度图解耦协同过滤 Deep Disentangled Collaborative Filtering with Graph Global Information 计算机科学, 2023, 50(1): 41-51. https://doi.org/10.11896/jsjkx.220900255
[2]	顾希之, 邵蓥侠. 基于影响力剪枝的图神经网络快速计算图精简 Fast Computation Graph Simplification via Influence-based Pruning for Graph Neural Network 计算机科学, 2023, 50(1): 52-58. https://doi.org/10.11896/jsjkx.220900032
[3]	孙凯丽, 罗旭东, 罗有容. 预训练语言模型的应用综述 Survey of Applications of Pretrained Language Models 计算机科学, 2023, 50(1): 176-184. https://doi.org/10.11896/jsjkx.220800223
[4]	周芳泉, 成卫青. 基于全局增强图神经网络的序列推荐 Sequence Recommendation Based on Global Enhanced Graph Neural Network 计算机科学, 2022, 49(9): 55-63. https://doi.org/10.11896/jsjkx.210700085
[5]	闫佳丹, 贾彩燕. 基于双图神经网络信息融合的文本分类方法 Text Classification Method Based on Information Fusion of Dual-graph Neural Network 计算机科学, 2022, 49(8): 230-236. https://doi.org/10.11896/jsjkx.210600042
[6]	齐秀秀, 王佳昊, 李文雄, 周帆. 基于概率元学习的矩阵补全预测融合算法 Fusion Algorithm for Matrix Completion Prediction Based on Probabilistic Meta-learning 计算机科学, 2022, 49(7): 18-24. https://doi.org/10.11896/jsjkx.210600126
[7]	杨炳新, 郭艳蓉, 郝世杰, 洪日昌. 基于数据增广和模型集成策略的图神经网络在抑郁症识别上的应用 Application of Graph Neural Network Based on Data Augmentation and Model Ensemble in Depression Recognition 计算机科学, 2022, 49(7): 57-63. https://doi.org/10.11896/jsjkx.210800070
[8]	熊中敏, 舒贵文, 郭怀宇. 融合用户偏好的图神经网络推荐模型 Graph Neural Network Recommendation Model Integrating User Preferences 计算机科学, 2022, 49(6): 165-171. https://doi.org/10.11896/jsjkx.210400276
[9]	邓朝阳, 仲国强, 王栋. 基于注意力门控图神经网络的文本分类 Text Classification Based on Attention Gated Graph Neural Network 计算机科学, 2022, 49(6): 326-334. https://doi.org/10.11896/jsjkx.210400218
[10]	余皑欣, 冯秀芳, 孙静宇. 结合物品相似性的社交信任推荐算法 Social Trust Recommendation Algorithm Combining Item Similarity 计算机科学, 2022, 49(5): 144-151. https://doi.org/10.11896/jsjkx.210300217
[11]	李勇, 吴京鹏, 张钟颖, 张强. 融合快速注意力机制的节点无特征网络链路预测算法 Link Prediction for Node Featureless Networks Based on Faster Attention Mechanism 计算机科学, 2022, 49(4): 43-48. https://doi.org/10.11896/jsjkx.210800276
[12]	曹合心, 赵亮, 李雪峰. 图神经网络在Text-to-SQL解析中的技术研究 Technical Research of Graph Neural Network for Text-to-SQL Parsing 计算机科学, 2022, 49(4): 110-115. https://doi.org/10.11896/jsjkx.210200173
[13]	苗旭鹏, 周跃, 邵蓥侠, 崔斌. GSO:基于图神经网络的深度学习计算图子图替换优化框架 GSO:A GNN-based Deep Learning Computation Graph Substitutions Optimization Framework 计算机科学, 2022, 49(3): 86-91. https://doi.org/10.11896/jsjkx.210700199
[14]	丁锋, 孙晓. 基于注意力机制和BiLSTM-CRF的消极情绪意见目标抽取 Negative-emotion Opinion Target Extraction Based on Attention and BiLSTM-CRF 计算机科学, 2022, 49(2): 223-230. https://doi.org/10.11896/jsjkx.210100046
[15]	杨旭华, 金鑫, 陶进, 毛剑飞. 基于图神经网络和依存句法分析的文本分类 Text Classification Based on Graph Neural Networks and Dependency Parsing 计算机科学, 2022, 49(12): 293-300. https://doi.org/10.11896/jsjkx.220300195

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed