Please wait a minute...
Advanced Search
数据分析与知识发现  2021, Vol. 5 Issue (9): 97-106     https://doi.org/10.11925/infotech.2096-3467.2021.0146
     研究论文 本期目录 | 过刊浏览 | 高级检索 |
基于图卷积神经网络和依存句法分析的网民负面情感分析研究*
范涛1,王昊1(),吴鹏2
1南京大学信息管理学院 南京 210023
2南京理工大学经济管理学院 南京 210094
Sentiment Analysis of Online Users' Negative Emotions Based on Graph Convolutional Network and Dependency Parsing
Fan Tao1,Wang Hao1(),Wu Peng2
1School of Information Management, Nanjing University, Nanjing 210023, China
2School of Economics and Management, Nanjing University of Science and Technology, Nanjing 210094, China
全文: PDF (1390 KB)   HTML ( 21
输出: BibTeX | EndNote (RIS)      
摘要 

【目的】 探索结合网络舆情文本的语义特征和依存结构信息进行网民负面情感分析研究。【方法】提出基于图卷积神经网络和依存句法分析的网民负面情感分析模型。模型结合双向长短期记忆网络和自注意力机制抽取文本特征作为依存句法图中的节点特征,应用图卷积神经网络对生成的节点特征和依存句法图对应的邻接矩阵进行训练学习,输出负面情感类别(愤怒、厌恶、恐惧和悲伤)。【结果】结合新冠疫情等网络舆情数据进行实证研究,并与相关基线模型作比较。实验结果表明,所提模型具有一定的优越性,在“恐惧”这一情感类别中,识别准确率达到 93.535 %【局限】 所提模型仅在网络舆情数据集中进行测试,未在公开数据集中进一步验证。【结论】依存句法结构信息的加入以及图卷积神经网络和注意力机制的应用能够有效提升模型的负面情感分析能力。

服务
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章
范涛
王昊
吴鹏
关键词 网络舆情负面情感图卷积神经网络自注意力机制依存句法    
Abstract

[Objective] This paper develop news method to improve the negative sentiment analysis of online users. [Objective] We proposed a model based on Graph Convolutional Networks (GCN) and dependency parsing. This model combined the BiLSTM and attention mechanism to extract textual features, which were then used as the vertex features. Third, we utilized the GCN to train the vertex features and the corresponding adjacency matrices. Finally, the model generated four types of emotions (anger, disgust, fear and sadness). [Results] We conducted an empirical study with online public opinion datasets (i.e., “COVID-19”) and compared the performance of our model with the baseline models. We found that the proposed model has certain advantages. For the emotion of “fear”, the recognition accuracy reached 93.535 %. [Limitations] We only examined the proposed model with online public opinion datasets. More research is needed to evaluate its performance with other public datasets. [Conclusions] Combining the dependency parsing information, the GCN, and the attention mechanism could increase the performance of negative sentiment analysis.

Key wordsOnline Public Opinion    Negative Emotions    Graph Convolutional Network    Self-Attention    Dependency Parsing
收稿日期: 2021-02-11      出版日期: 2021-10-15
ZTFLH:  分类号: G202  
基金资助:*国家自然科学基金项目(71774084);南京大学文科青年跨学科团队专项的研究成果之一(010814370113)
通讯作者: 王昊     E-mail: ywhaowang@nju.edu.cn
引用本文:   
范涛,王昊,吴鹏. 基于图卷积神经网络和依存句法分析的网民负面情感分析研究*[J]. 数据分析与知识发现, 2021, 5(9): 97-106.
Fan Tao,Wang Hao,Wu Peng. Sentiment Analysis of Online Users' Negative Emotions Based on Graph Convolutional Network and Dependency Parsing. Data Analysis and Knowledge Discovery, 2021, 5(9): 97-106.
链接本文:  
https://manu44.magtech.com.cn/Jwk_infotech_wk3/CN/10.11925/infotech.2096-3467.2021.0146      或      https://manu44.magtech.com.cn/Jwk_infotech_wk3/CN/Y2021/V5/I9/97
Fig.1  依存句法图
Fig.2  基于图卷积神经网络和依存句法分析的网民负面情感分析模型
Fig.3  依存句法树及其对应的邻接矩阵示例
情感 数量
非负面 22 536
负面 愤怒 6 839
厌恶 6 645
恐惧 5 008
悲伤 5 584
Table 1  情感标注结果
模型 Acc / % Macro _ P / % Macro _ R / % Macro _ F 1 / %
BiLSTM-Attention-GCN 84 . 842 84 . 941 85 . 143 84 . 937
BiLSTM-GCN 82.766 82.879 82.851 82.592
BiLSTM-attention 81.042 80.946 80.951 80.923
BiLSTM 79.610 79.749 79.085 79.219
CNN 77.990 78.416 77.042 77.347
Logistic 73.173 73.041 73.868 72.981
Table 2  网民负面情感分析结果
Fig.4  不同模型的情感分类混淆矩阵
Fig.5  隐藏层单元数对模型性能的影响
Fig.6  GCN层数对模型性能的影响
模型 Acc / % Marco _ P / % Marco _ R / % Marco _ F 1 / %
BiLSTM-Attention-GCN 82 . 667 82 . 657 82 . 68 82 . 665
BiLSTM-GCN 82.098 82.089 82.058 82.070
BiLSTM-Attention 81.712 81.723 81.650 81.673
BiLSTM 81.401 81.409 81.340 81.362
CNN 79.835 79.814 79.807 79.810
Logistic 74.225 74.197 74.187 74.192
Table 3  网民情感二分类结果
[1] 吴鹏, 强韶华, 高庆宁. 基于SOAR模型的网民群体负面情感建模研究[J]. 中国管理科学, 2018, 26(3):126-138.
[1] ( Wu Peng, Qiang Shaohua, Gao Qingning. Modelling Internet Users' Negative Emotion Based on SOAR Model[J]. Chinese Journal of Management Science, 2018, 26(3):126-138.)
[2] 吴鹏, 应杨, 沈思. 基于双向长短期记忆模型的网民负面情感分类研究[J]. 情报学报, 2018, 37(8):845-853.
[2] ( Wu Peng, Ying Yang, Shen Si. Negative Emotions of Online Users' Analysis Based on Bidirectional Long Short-Term Memory[J]. Journal of the China Society for Scientific and Technical Information, 2018, 37(8):845-853.)
[3] 邓君, 孙绍丹, 王阮, 等. 基于Word2Vec和SVM的微博舆情情感演化分析[J]. 情报理论与实践, 2020, 43(8):112-119.
[3] ( Deng Jun, Sun Shaodan, Wang Ruan, et al. Evolution Analysis of Weibo Public Opinion Emotion Based on Word2Vec and SVM[J]. Information Studies: Theory & Application, 2020, 43(8):112-119.)
[4] Hou X C, Huang J, Wang G T, et al. Selective Attention Based Graph Convolutional Networks for Aspect-Level Sentiment Classification[OL]. arXiv Preprint,arXiv: 1910. 10857.
[5] 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[J]. Knowledge-Based Systems, 2020, 205:106292.
doi: 10.1016/j.knosys.2020.106292
[6] Wang R, Xin X, Chang W, et al. Chinese NER with Height-Limited Constituent Parsing [C]//Proceedings of the AAAI Conference on Artificial Intelligence. 2019: 7160-7167.
[7] 徐冰冰, 岑科廷, 黄俊杰, 等. 图卷积神经网络综述[J]. 计算机学报, 2020, 43(5):755-780.
[7] ( Xu Bingbing, Cen Keting, Huang Junjie, et al. A Survey on Graph Convolutional Neural Network[J]. Chinese Journal of Computers, 2020, 43(5):755-780.)
[8] Kipf T N, Welling M. Semi-Supervised Classification with Graph Convolutional Networks[OL]. arXiv Preprint, arXiv: 1609.02907.
[9] Zhao P L, Hou L L, Wu O. Modeling Sentiment Dependencies with Graph Convolutional Networks for Aspect-level Sentiment Classification[J]. Knowledge-Based Systems, 2020, 193:105443.
doi: 10.1016/j.knosys.2019.105443
[10] Lai Y N, Zhang L F, Han D H, et al. Fine-Grained Emotion Classification of Chinese Microblogs Based on Graph Convolution Networks[J]. World Wide Web, 2020, 23(5):2771-2787.
doi: 10.1007/s11280-020-00803-0
[11] Zhang H, Goodfellow I, Metaxas D, et al. Self-Attention Generative Adversarial Networks [C]//Proceedings of the 36th International Conference on Machine Learning. 2019: 7354-7363.
[12] 孙靖超. 基于优化深度双向自编码网络的舆情情感识别研究[J]. 情报杂志, 2020, 39(6):159-163, 195.
[12] ( Sun Jingchao. Sentiment Analysis of Network Public Opinion Based on Optimized Bidirectional Encoder Representations from Transformers[J]. Journal of Intelligence, 2020, 39(6):159-163, 195.)
[13] 范涛, 吴鹏, 曹琪. 基于深度学习的多模态融合网民情感识别研究[J]. 信息资源管理学报, 2020, 10(1):39-48.
[13] ( Fan Tao, Wu Peng, Cao Qi. The Research of Sentiment Recognition of Online Users Based on DNNs Multimodal Fusion[J]. Journal of Information Resources Management, 2020, 10(1):39-48.)
[14] Chen D Q, Manning C. A Fast and Accurate Dependency Parser Using Neural Networks [C]//Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing(EMNLP). 2014: 740-750.
[15] Agarwal B, Poria S, Mittal N, et al. Concept-Level Sentiment Analysis with Dependency-Based Semantic Parsing: A Novel Approach[J]. Cognitive Computation, 2015, 7(4):487-499.
doi: 10.1007/s12559-014-9316-6
[16] Gómez-Rodríguez C, Alonso-Alonso I, Vilares D. How Important is Syntactic Parsing Accuracy? An Empirical Evaluation on Rule-Based Sentiment Analysis[J]. Artificial Intelligence Review, 2019, 52(3):2081-2097.
doi: 10.1007/s10462-017-9584-0
[17] 潘浩, 卫宇杰, 潘尔顺. 基于自动提取句法模板的情感分析[J]. 中文信息学报, 2019, 33(9):129-140.
[17] ( Pan Hao, Wei Yujie, Pan Ershun. Emotion Analysis Based on Automatic Extraction of Syntactic Patterns[J]. Journal of Chinese Information Processing, 2019, 33(9):129-140.)
[18] 王昊, 邓三鸿, 朱立平, 等. 大数据环境下政务数据的情报价值及其利用研究——以海关报关商品归类风险规避为例[J]. 科技情报研究, 2020, 2(4):74-89.
[18] ( Wang Hao, Deng Sanhong, Zhu Liping, et al. A Study of Intelligence Value and Employment of Political Data in Big Data Environment——The Risk Avoidance of Customs Declaration Commodities[J]. Scientific Information Research, 2020, 2(4):74-89.)
[19] He K M, Gkioxari G, Dollár P, et al. Mask R-CNN [C]//Proceedings of the IEEE International Conference on Computer Vision. 2017: 2980-2988.
[20] Wang S Y, Huang M L, Deng Z D. Densely Connected CNN with Multi-scale Feature Attention for Text Classification [C]//Proceedings of the 27th International Joint Conference on Artificial Intelligence. 2018: 4468-4474.
[21] Bastings J, Titov I, Aziz W, et al. Graph Convolutional Encoders for Syntax-aware Neural Machine Translation [C]//Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing. 2017: 1957-1967.
[22] Nguyen T H, Grishman R. Graph Convolutional Networks with Argument-Aware Pooling for Event Detection [C]//Proceedings of the AAAI Conference on Artificial Intelligence. 2018.
[23] Mikolov T, Chen K, Corrado G, et al. Efficient Estimation of Word Representations in Vector Space[OL]. arXiv Preprint, arXiv: 1301.3781.
[24] Gers F A, Schmidhuber J, Cummins F. Learning to Forget: Continual Prediction with LSTM[J]. Neural Computation, 2000, 12(10):2451-2471.
pmid: 11032042
[25] 吴鹏, 刘恒旺, 沈思. 基于深度学习和OCC情感规则的网络舆情情感识别研究[J]. 情报学报, 2017, 36(9):972-980.
[25] ( Wu Peng, Liu Hengwang, Shen Si. Sentiment Analysis of Network Public Opinion Based on Deep Learning and OCC[J]. Journal of the China Society for Scientific and Technical Information, 2017, 36(9):972-980.)
[26] Ekman P. An Argument for Basic Emotions[J]. Cognition and Emotion, 1992, 6(3-4):169-200.
doi: 10.1080/02699939208411068
[27] Tromp E, Pechenizkiy M. Rule-based Emotion Detection on Social Media: Putting Tweets on Plutchik's Wheel[OL]. arXiv Preprint, arXiv: 1412.4682.
[28] Che W X, Li Z H, Liu T. LTP: A Chinese Language Technology Platform [C]//Proceedings of the 23rd International Conference on Computational Linguistics: Demonstrations. 2010: 13-16.
[29] Xie J, Chen B, Gu X L, et al. Self-Attention-Based BiLSTM Model for Short Text Fine-Grained Sentiment Classification[J]. IEEE Access, 2019, 7:180558-180570.
doi: 10.1109/Access.6287639
[30] Xu G X, Meng Y T, Qiu X Y, et al. Sentiment Analysis of Comment Texts Based on BiLSTM[J]. IEEE Access, 2019, 7:51522-51532.
doi: 10.1109/Access.6287639
[31] Kim Y. Convolutional Neural Networks for Sentence Classification[OL]. arXiv Preprint, arXiv: 1408.5882.
[32] Prabhat A, Khullar V. Sentiment Classification on Big Data Using Naïve Bayes and Logistic Regression [C]//Proceedings of 2017 International Conference on Computer Communication and Informatics. 2017: 1-5.
[1] 韩普,张展鹏,张明淘,顾亮. 基于多特征融合的中文疾病名称归一化研究*[J]. 数据分析与知识发现, 2021, 5(5): 83-94.
[2] 程铁军, 王曼, 黄宝凤, 冯兰萍. 基于CEEMDAN-BP模型的突发事件网络舆情预测研究*[J]. 数据分析与知识发现, 2021, 5(11): 59-67.
[3] 邓建高,张璇,傅柱,韦庆明. 基于系统动力学的突发事件网络舆情传播研究:以“江苏响水爆炸事故”为例*[J]. 数据分析与知识发现, 2020, 4(2/3): 110-121.
[4] 梁艳平,安璐,刘静. 同类突发公共卫生事件微博话题共振研究*[J]. 数据分析与知识发现, 2020, 4(2/3): 122-133.
[5] 丁晟春,俞沣洋,李真. 网络舆情潜在热点主题识别研究*[J]. 数据分析与知识发现, 2020, 4(2/3): 29-38.
[6] 黄微,赵江元,闫璐. 网络热点事件话题漂移指数构建与实证研究*[J]. 数据分析与知识发现, 2020, 4(11): 92-101.
[7] 李博诚,张云秋,杨铠西. 面向微博商品评论的情感标签抽取研究 *[J]. 数据分析与知识发现, 2019, 3(9): 115-123.
[8] 梅妍霜,朱恒民,魏静. 媒体协同对网络舆情扩散的作用机制研究*[J]. 数据分析与知识发现, 2019, 3(2): 65-71.
[9] 严娇,马静,房康. 基于融合共现距离的句法网络下文本语义相似度计算 *[J]. 数据分析与知识发现, 2019, 3(12): 93-100.
[10] 贾隆嘉, 张邦佐. 高校网络舆情安全中主题分类方法研究*——以新浪微博数据为例[J]. 数据分析与知识发现, 2018, 2(7): 55-62.
[11] 李琳, 李辉. 一种基于概念向量空间的文本相似度计算方法[J]. 数据分析与知识发现, 2018, 2(5): 48-58.
[12] 王璟琦, 李锐, 吴华意. 基于空间自相关的网络舆情话题演化时空规律分析*[J]. 数据分析与知识发现, 2018, 2(2): 64-73.
[13] 李真, 丁晟春, 王楠. 网络舆情观点主题识别研究*[J]. 数据分析与知识发现, 2017, 1(8): 18-30.
[14] 王晰巍, 张柳, 李师萌, 王楠阿雪. 新媒体环境下社会公益网络舆情传播研究* ——以新浪微博“画出生命线”话题为例[J]. 数据分析与知识发现, 2017, 1(6): 93-101.
[15] 丁晟春,龚思兰,李红梅. 基于突发主题词和凝聚式层次聚类的微博突发事件检测研究*[J]. 现代图书情报技术, 2016, 32(7-8): 12-20.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
版权所有 © 2015 《数据分析与知识发现》编辑部
地址:北京市海淀区中关村北四环西路33号 邮编:100190
电话/传真:(010)82626611-6626,82624938
E-mail:jishu@mail.las.ac.cn