基于双重多视角表示的目标级隐性情感分类

doi:10.19678/j.issn.1000-3428.0066459

摘要/Abstract

摘要：

目标级隐性情感分类是自然语言处理中一项重要的情感分析任务。目前多数研究主要侧重于对上下文感知的目标进行建模，且建模信息源较为单一，难以充分捕获到目标词在文本中的隐性情感。针对该问题，提出基于双重多视角表示学习的目标级隐性情感分类方法，采用3种视角对目标和输入文本进行建模，分别设计文本自身的表示学习、图视角下的表示学习以及外部知识视角下的表示学习，并通过卷积神经网络将3种视角下的表示进行深度融合。此外，同时采用上述3种视角对目标进行表示学习，将文本的语义表示和目标的语义表示相结合，并输入到情感极性分类器中。在5个公共数据集上进行实验并与8个基线模型的对比结果表明，该方法性能达到了最优水平，在NewsMTSC-mt和NewsMTSC-rw隐性情感分析数据集上的$ \mathrm{F}{1}_{\mathrm{m}} $值分别比最好模型提高1.0%和2.6%，在Laptop14、Restaurant14和Twitter显性情感分析数据集上的$ \mathrm{F}{1}_{\mathrm{m}} $值分别比最好模型提高3.6%、1.4%和1.6%。

关键词: 目标级隐性情感分类, 自然语言处理, 情感分析, 双重多视角, 表示学习

Abstract:

Target-level implicit sentiment classification is a critical sentiment analysis task in natural language processing. Many existing studies mainly focused on modeling context-aware targets, and their modeling information source were relatively single, making it difficult to adequately capture the implicit sentiment of the target in the text. This study proposes a target-level sentiment classification method based on dual multiview representation learning that models the target and text from three information views. Specifically, this study designs a representation learning model from the text, the view of the graph, and the view of external knowledge and exploits a convolutional neural network to deeply integrate the representations of the three views. Moreover, the proposed method learns target-dependent representations from these views. Finally, the semantic representations of the text and the target are combined and fed into the sentiment classifier. The results of experiments conducted on five public datasets and comparative experiments with eight baseline models show that the solution achieves state-of-the-art performance. In particular, the F1_m of the proposed model is 1.0% and 2.6% higher than those of previous best models on NewsMTSC-mt and NewsMTSC-rw implicit sentiment analysis datasets, respectively. In addition, the F1_m of the proposed model is 3.6%, 1.4%, and 1.6% higher than those of the previous best models on Laptop14, Restaurant14, and Twitter explicit emotion analysis datasets, respectively.

Key words: target-level implicit sentiment classification, natural language processing, sentiment analysis, dual multiview, representation learning

崔蒙蒙, 刘井平, 阮彤, 宋雨秋, 杜渂. 基于双重多视角表示的目标级隐性情感分类[J]. 计算机工程, 2024, 50(1): 79-90.

Mengmeng CUI, Jingping LIU, Tong RUAN, Yuqiu SONG, Wen DU. Target-Level Implicit Sentiment Classification Based on Dual Multiview Representation[J]. Computer Engineering, 2024, 50(1): 79-90.

http://www.ecice06.com/CN/Y2024/V50/I1/79

图/表 15

图1 DMVRL整体结构

Fig.1 The overall structure of DMVRL

图2 在5个数据集上不同窗口大小下模型的性能比较

Fig.2 Performance comparison of models with different window size on five datasets

图3 目标词和先验情感词之间的注意力分数可视化

Fig.3 The visualization of attention scores between target words and pre-defined prior sentiment words

参考文献 35

1	韩忠明, 李梦琪, 刘雯, 等. 网络评论方面级观点挖掘方法研究综述. 软件学报, 2018, 29(2): 417- 441.
	HAN Z M, LI M Q, LIU W, et al. Survey of studies on aspect-based opinion mining of Internet. Journal of Software, 2018, 29(2): 417- 441.
2	JIANG L, YU M, ZHOU M, et al. Target-dependent Twitter sentiment classification[C]//Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics. New York, USA: ACM Press, 2011: 151-160.
3	邱俊安. 中文隐喻的识别与情感分析[D]. 武汉: 武汉理工大学, 2021.
	QIU J A. Recognition and emotion analysis of Chinese metaphors[D]. Wuhan: Wuhan University of Technology, 2021. (in Chinese)
4	陈慧敏, 金思辰, 林微, 等. 新冠疫情相关社交媒体谣言传播量化分析. 计算机研究与发展, 2021, 58(7): 1366- 1384.
	CHEN H M, JIN S C, LIN W, et al. Quantitative analysis of the spread of social media rumors related to the new crown epidemic. Computer Research and Development, 2021, 58(7): 1366- 1384.
5	CHEN H Y, CHEN H H. Implicit polarity and implicit aspect recognition in opinion mining[C]//Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics. Berlin, Germany: Springer, 2016: 20-25.
6	DING X W, LIU B. The utility of linguistic rules in opinion mining[C]//Proceedings of the 30th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval. New York, USA: ACM Press, 2007: 811-812.
7	DING X W, LIU B, YU P S. A holistic lexicon-based approach to opinion mining[C]//Proceedings of 2008 International Conference on Web Search and Data Mining. New York, USA: ACM Press, 2008: 231-240.
8	YANG H, ZENG B Q. Enhancing fine-grained sentiment classification exploiting local context embedding[EB/OL]. [2022-11-01]. https://arxiv.org/abs/2010.00767.pdf.
9	SONG Y W, WANG J H, JIANG T, et al. Attentional encoder network for targeted sentiment classification[EB/OL]. [2022-11-01]. https://arxiv.org/abs/1902.09314.pdf.
10	LI X, BING L D, LAM W, et al. Transformation networks for target-oriented sentiment classification[C]//Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics. Washington D. C., USA: IEEE Press, 2018: 946-956.
11	JOHNSON R, ZHANG T. Semi-supervised convolutional neural networks for text categorization via region embedding. Neural Information Processing Systems, 2015, 28, 919- 927.
12	MILLER G A. WordNet. Communications of the ACM, 1995, 38(11): 39- 41. doi: 10.1145/219717.219748
13	CAMBRIA E, PORIA S, HAZARIKA D, et al. SenticNet 5: discovering conceptual primitives for sentiment analysis by means of context embeddings. Artificial Intelligence, 2018, 32(1): 328- 337.
14	SPEER R, CHIN J, HAVASI C. ConceptNet 5.5: an open multilingual graph of general knowledge. Artificial Intelligence, 2017, 31(1): 4444- 4451.
15	HU M Q, LIU B. Mining and summarizing customer reviews[C]//Proceedings of the 10th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York, USA: ACM Press, 2004: 168-177.
16	MOHAMMAD S M, TURNEY P D. Emotions evoked by common words and phrases: using mechanical Turk to create an emotion lexicon[C]//Proceedings of 2010 Workshop on Computational Approaches to Analysis and Generation of Emotion in Text. New York, USA: ACM Press, 2010: 26-34.
17	WILSON T, WIEBE J, HOFFMANN P. Recognizing contextual polarity in phrase-level sentiment analysis[C]//Proceedings of Conference on Human Language Technology and Empirical Methods in Natural Language Processing. New York, USA: ACM Press, 2005: 347-354.
18	彭祝亮, 刘博文, 范程岸, 等. 基于BLSTM与方面注意力模块的情感分类方法. 计算机工程, 2020, 46(3): 60-65, 72. URL
	PENG Z L, LIU B W, FAN C A, et al. Sentiment classification method based on BLSTM and aspect attention module. Computer Engineering, 2020, 46(3): 60-65, 72. URL
19	MA D H, LI S J, ZHANG X D, et al. Interactive attention networks for aspect-level sentiment classification[EB/OL]. [2022-11-01]. https://arxiv.org/abs/1709.00893.pdf.
20	ZENG B Q, YANG H, XU R Y, et al. LCF: a local context focus mechanism for aspect-based sentiment classification. Applied Sciences, 2019, 9(16): 3389.
21	TANG D, QIN B, FENG X, et al. Effective LSTMs for target-dependent sentiment classification[C]//Proceedings of the 26th International Conference on Computational Linguistics. Washington D. C., USA: IEEE Press, 2016: 3298-3307.
22	SUN K, ZHANG R C, MENSAH S, et al. Aspect-level sentiment analysis via convolution over dependency tree[C]//Proceedings of 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing. Stroudsburg, USA: Association for Computational Linguistics, 2019: 5679-5688.
23	ZHANG C, LI Q C, SONG D W. Aspect-based sentiment classification with aspect-specific graph convolutional networks[EB/OL]. [2022-11-01]. https://arxiv.org/abs/1909.03477.pdf.
24	ZHAO P L, HOU L L, WU O. Modeling sentiment dependencies with graph convolutional networks for aspect-level sentiment classification. Knowledge-Based Systems, 2020, 193, 105443.
25	MA Y K, PENG H Y, CAMBRIA E. Targeted aspect-based sentiment analysis via embedding commonsense knowledge into an attentive LSTM. Artificial Intelligence, 2018, 32(1): 548- 556.
26	HAMBORG F, DONNAY K. NewsMTSC: a dataset for (multi-)target-dependent sentiment classification in political news articles[C]//Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics. Berlin, Germany: Springer, 2021: 257-268.
27	SHI X J, CHEN Z R, WANG H, et al. Convolutional LSTM network: a machine learning approach for precipitation nowcasting[EB/OL]. [2022-11-01]. https://arxiv.org/abs/1506.04214.pdf.
28	DEVLIN J, CHANG M W, LEE K, et al. BERT: pre-training of deep bidirectional transformers for language understanding[EB/OL]. [2022-11-01]. https://arxiv.org/abs/1810.04805.pdf.
29	LIU Z A, LIN W, SHI Y, et al. A robustly optimized BERT pre-training approach with post-training. Berlin, Germany: Springer, 2021.
30	ZHOU J, HUANG J X, HU Q V, et al. SK-GCN: modeling syntax and knowledge via graph convolutional network for aspect-level sentiment classification. Knowledge-Based Systems, 2020, 205, 106292.
31	PONTIKI M, GALANIS D, PAPAGEORGIOU H, et al. SemEval-2015 task 12: aspect based sentiment analysis[C]//Proceedings of the 9th International Workshop on Semantic Evaluation. Stroudsburg, USA: Association for Computational Linguistics, 2015: 3325-3338.
32	DONG L, WEI F R, TAN C Q, et al. Adaptive recursive neural network for target-dependent twitter sentiment classification[C]//Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics. Stroudsburg, USA: Association for Computational Linguistics, 2014: 49-54.
33	SUN C, HUANG L Y, QIU X P. Utilizing BERT for aspect-based sentiment analysis via constructing auxiliary sentence[EB/OL]. [2022-11-01]. https://arxiv.org/abs/1903.09588.pdf.
34	GAO Z J, FENG A, SONG X Y, et al. Target-dependent sentiment classification with BERT. IEEE Access, 2019, 7, 154290- 154299.
35	LI R, CHEN H, FENG F, et al. Dual graph convolutional networks for aspect-based sentiment analysis[C]//Proceedings of IEEE IJCNLPʼ21. Washington D. C., USA: IEEE Press, 2021: 356-367.

[1]	隋国华, 李陶然, 刘昊, 陈林, 汪卫. 基于图表示学习的领域知识图谱推理技术研究[J]. 计算机工程, 2023, 49(9): 89-98.
[2]	李鸿鹏, 马博, 杨雅婷, 王磊, 王震, 李晓. 基于槽位语义增强提示学习的篇章级事件抽取方法[J]. 计算机工程, 2023, 49(9): 23-31.
[3]	郭艳霞, 金勇, 唐宏, 彭金枝. 基于动态卷积与残差门控的多模态情感识别[J]. 计算机工程, 2023, 49(7): 94-101.
[4]	代祖华, 刘园园, 狄世龙. 语义增强的图神经网络方面级文本情感分析[J]. 计算机工程, 2023, 49(6): 71-80.
[5]	陈昱瑾, 王晶, 武志昊, 赵耀帅, 林友芳. 基于图卷积网络融合群组关系的社会化推荐方法[J]. 计算机工程, 2023, 49(5): 112-121.
[6]	李静雯, 赵奎. 基于改进PCFG算法的口令猜测方法[J]. 计算机工程, 2023, 49(5): 38-47.
[7]	沈学利, 韩倩雯. 基于注意力机制的场感知点击率预测模型[J]. 计算机工程, 2023, 49(3): 80-86,94.
[8]	杨文忠, 丁甜甜, 康鹏, 卜文秀. 基于舆情新闻的中文关键词抽取综述[J]. 计算机工程, 2023, 49(3): 1-17.
[9]	蔡瑞初, 张盛强, 许柏炎. 基于结构感知混合编码模型的代码注释生成方法[J]. 计算机工程, 2023, 49(2): 61-69.
[10]	袁立宁, 胡皓, 刘钊. 基于多通道图卷积自编码器的图表示学习[J]. 计算机工程, 2023, 49(2): 150-160,174.
[11]	王春东, 孙嘉琪, 杨文军. 基于矫正理解的中文文本对抗样本生成方法[J]. 计算机工程, 2023, 49(2): 37-45.
[12]	富坤, 孙明磊, 郝玉涵, 刘赢华. 基于对抗训练的伪标签约束自编码器[J]. 计算机工程, 2023, 49(11): 123-130.
[13]	吴奇林, 党亚固, 熊山威, 吉旭, 毕可鑫. 基于混合特征网络的学生评教文本情感分析模型[J]. 计算机工程, 2023, 49(11): 24-29, 39.
[14]	杨海洋, 张兴鹏. 融合多通道图卷积网络的方面级情感分析模型[J]. 计算机工程, 2023, 49(11): 61-69.
[15]	孔博, 韩虎, 陈景景, 白雪, 邓飞. 基于虚拟依存关系与知识增强的方面级情感分析[J]. 计算机工程, 2023, 49(10): 53-63.

选择文件类型/文献管理软件名称

选择包含的内容