基于深度BiLSTM和图卷积网络的方面级情感分析

计算机工程与科学 ›› 2022, Vol. 44 ›› Issue (10): 1893-1900.

• 人工智能与数据挖掘 • 上一篇

基于深度BiLSTM和图卷积网络的方面级情感分析

杨春霞1,2,3，宋金剑1,2,3，姚思诚1,2,3

（1.南京信息工程大学自动化学院，江苏南京210044；2.江苏省大数据分析技术重点实验室，江苏南京 210044；
3.江苏省大气环境与装备技术协同创新中心，江苏南京 210044）

收稿日期:2021-03-19 修回日期:2021-05-28 接受日期:2022-10-25 出版日期:2022-10-25 发布日期:2022-10-28
基金资助:
国家自然科学基金（51705260,61273229）

Aspect-level sentiment analysis based on deep BiLSTM and graph convolutional networks

YANG Chun-xia1,2,3,SONG Jin-jian1,2,3,YAO Si-cheng1,2,3

(1.School of Automation,Nanjing University of Information Science & Technology,Nanjing 210044;
2.Jiangsu Key Laboratory of Big Data Analysis Technology(B-DAT),Nanjing 210044;
3.Jiangsu Collaborative Innovation Center of Atmospheric Environment and
Equipment Technology(CICAEET),Nanjing 210044,China)

Received:2021-03-19 Revised:2021-05-28 Accepted:2022-10-25 Online:2022-10-25 Published:2022-10-28

摘要/Abstract

摘要： 现有方面级情感分析方法，存在无法获取最优文本表示和使用普通图卷积网络不能提取依存图中深层结构信息的问题。为此，提出了一种基于深度BiLSTM（DBiLSTM）和紧密连接的图卷积网络（DDGCN）模型。首先,通过DBiLSTM获取方面词与上下文单词间的深层语义信息；其次，在原始图卷积网络中加入紧密连接，以生成能提取深层结构信息的紧密图卷积网络；然后，利用改进后的图卷积网络捕获依存图上的结构信息；最终，将融合2种深层信息的文本表示用于情感分类。3个数据集上的实验结果表明，DDGCN模型相比对比模型在准确度和F1上均有提升。

关键词: 方面级情感分析, 图卷积网络, 深度BiLSTM, 紧密连接

Abstract: In the existing aspect-level sentiment analysis methods, there are problems that the optimal text representation cannot be obtained, and the general graph convolutional network cannot be used to extract the deep structural information in the dependency graph. For these problems, this paper proposes a graph convolutional network (DDGCN) model based on deep BiLSTM and dense connections. Firstly, the deep semantic information between aspect words and context words is obtained by deep BiLSTM. Secondly, dense connections are added to the original graph convolutional network to generate a dense graph convolutional network that can extract the deep structural information. Then, the improved graph convolutional network is used to capture the structural information on the dependency graph. Finally, the text representation combining the two deep information is used for sentiment classification. Experimental results on three datasets show that DDGCN model improves both accuracy and F1 compared with the comparative model.

Key words: aspect-level sentiment analysis, graph convolutional network, deep BiLSTM, dense connection

杨春霞, 宋金剑, 姚思诚, . 基于深度BiLSTM和图卷积网络的方面级情感分析[J]. 计算机工程与科学, 2022, 44(10): 1893-1900.

YANG Chun-xia, SONG Jin-jian, YAO Si-cheng, . Aspect-level sentiment analysis based on deep BiLSTM and graph convolutional networks[J]. Computer Engineering & Science, 2022, 44(10): 1893-1900.

参考文献［9］

［1］	Pang B,Lee L L. Opinion mining and sentiment analysis［J］.Foundations and Trends in Information Retrieval,2008,2(1-2):1-135.
［2］	Wang K,Shen W Z,Yang Y Y,et al.Relational graph attention network for aspect-based sentiment analysis［J］.arXiv:2004.12362,2020.
［3］	Sun K,Zhang R C,Mensah S,et al.Aspect-level sentiment analysis via convolution over dependency tree［C］∥Proc of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing,2019:5683-5692.
［4］	Kipf T N, Welling M.Semi-supervised classification with graph convolutional networks［J］.arXiv:1609.02907,2016.
［5］	Han Zhe,Jiang Jing-fei,Qiao Lin-bo,et al.Design and implementation of event extraction model and accelerator based on FPGA［J］.Computer Engineering & Science,2020,42(11):1941-1948.(in Chinese)
［6］	Chen Jin-yin,Xu Xuan-heng,Wu Yang-yang,et al.Dynamic link prediction based on spatio temporal attention deep model［J］.Journal of Chinese Computer Systems,2019,40(11): 2365-2373.(in Chinese)
［7］	Li Q M,Han Z C,Wu X M.Deeper insights into graph convolutional networks for semi-supervised learning［C］∥Proc of the 32nd AAAI Conference on Artificial Intelligence,2018: 3538-3545.
［8］	Huang G,Liu Z,van der Maaten L,et al.Densely connected convolutional networks［C］∥Proc of the IEEE Conference on Computer Vision and Pattern Recognition,2017: 4700-4708.
［9］	Zeng Zheng,Li Li,Chen Jing.Deeply hierarchical bidirectional LSTM for sentiment classification ［J］.Computer Science,2018,45(8):213-217.(in Chinese)
［10］	Zhang C,Li Q C,Song D W.Aspect-based sentiment classification with aspect-specific graph convolutional networks［C］∥Proc of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing,2019:4560-4570.
［11］	Xu K Y L,Li C T,Tian Y L,et al.Representation learning on graphs with jumping knowledge networks［C］∥Proc of International Conference on Machine Learning,2018: 5453-5462.
［12］	Pennington J,Socher R,Manning C D.Glove: Global vectors for word representation［C］∥Proc of the 2014 Confe- rence on Empirical Methods in Natural Language Processing,2014:1532-1543.
［13］	Ma D H,Li S J,Zhang X D,et al.Interactive attention networks for aspect-level sentiment classification［C］∥Proc of the 26th International Joint Conference on Artificial Intelligence,2017: 4068-4074.
［14］	Li X,Bing L D,Lam W,et al.Transformation networks for target-oriented sentiment classification［C］∥Proc of the 56th Annual Meeting of the Association for Computational Linguistics,2018:946-956.
［15］	Song Y W,Wang J H,Jiang T,et al.Attentional encoder network for targeted sentiment classification［J］.arXiv:1902.09314,2019.
	附中文参考文献：
［5］	韩哲,姜晶菲,乔林波,等.基于FPGA的事件抽取模型与加速器的设计实现［J］.计算机工程与科学,2020,42(11):1941-1948.
［6］	陈晋音,徐轩桁,吴洋洋,等.基于时空注意力深度模型的动态链路预测［J］.小型微型计算机系统,2019,40(11): 2365-2373.
［9］	曾蒸，李莉，陈晶.用于情感分类的双向深度LSTM［J］.计算机科学,2018,45(8):213-217.

[1]	赵艺. 基于路径签名的改进时空图卷积网络[J]. 计算机工程与科学, 2022, 44(12): 2213-2219.
[2]	杨春霞, 姚思诚, 宋金剑, . 基于词共现的方面级情感分析模型[J]. 计算机工程与科学, 2022, 44(11): 2071-2079.
[3]	顾军华, 佘士耀, 樊帅, 张素琪. 基于用户长短期兴趣与知识图卷积网络的推荐[J]. 计算机工程与科学, 2021, 43(03): 511-517.

基于深度BiLSTM和图卷积网络的方面级情感分析

Aspect-level sentiment analysis based on deep BiLSTM and graph convolutional networks

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献［9］

相关文章 3

编辑推荐

Metrics

本文评价

基于深度BiLSTM和图卷积网络的方面级情感分析

Aspect-level sentiment analysis based on deep BiLSTM and graph convolutional networks

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献 ［9］

相关文章 3

编辑推荐

Metrics

本文评价

参考文献［9］