基于双通道交互的图卷积网络方面级情感分析
Aspect-Level Sentiment Analysis Based on Bi-Channel Interactive Graph Convolutional Network
摘要: 基于方面的情感分析旨在识别文本中特定方面的情感极性，目前，大多数研究将句法依赖树和图卷积神经网络应用于方面级情感分析，并取得了不错的结果。然而，当一个评论文本包含多个方面时，大多数方法对每个方面单独建模，从而忽略了方面词之间的情感联系。为了解决这一问题，本文提出了一种用于方面级情感分析的双通道交互式图卷积网络(BC-GCN)模型。该模型同时考虑了句子的句法结构信息以及多个方面的情感依赖性，并使用图卷积网络来学习其节点信息表示。特别是，为了更好地捕获方面词和观点词的表示，我们利用交互注意力机制来学习图卷积网络产生的句法信息特征和多方面情感依赖特征。在多组公开数据集上的实验结果表明，我们提出的改进方法能显著提高模型的性能。
Abstract: Aspect-based sentiment analysis aims to identify the emotional polarity of specific aspects of text, and currently, most studies apply syntactically dependent trees and graph convolutional neural networks to aspect-level sentiment analysis with good results. However, when a review text con-tains multiple aspects, most methods model each aspect individually, ignoring the emotional connection between aspect words. To solve this problem, a Bi-channel interactive graph convolutional network (BC-GCN) model for aspect-based sentiment analysis is proposed. The model considers both the syntactic structure information of sentences and the emotional dependence of multiple aspects, and uses the graph convolutional network to learn its node information representation. In particu-lar, in order to better capture the representation of aspect words and opinion words, we use the interactive attention mechanism to learn the syntactic information features and multifaceted emotional dependency features generated by graph convolutional networks. Experimental results on multiple sets of public datasets show that the proposed improved method can significantly improve the performance of the model.
文章引用：信雪晴, 单菁, 王佳英. 基于双通道交互的图卷积网络方面级情感分析[J]. 计算机科学与应用, 2022, 12(12): 2863-2874. https://doi.org/10.12677/CSA.2022.1212291

参考文献

[1] Han, Z.M., Li, M.Q., Liu, W., et al. (2018) Survey of Studies on Aspect-Based Opinion Mining of Internet. Journal of Software, 29, 417-441.
[2] 李胜旺, 杨艺, 许云峰, 张妍. 文本方面级情感分类方法综述[J]. 河北科技大学学报, 2020, 41(6): 518-527
[3] 张璞, 李逍, 刘畅. 基于规则的评价搭配抽取方法[J]. 计算机工程, 2019, 45(8): 217-223
[4] Li, P., Zhao, F., Li, Y. and Zhu, Z. (2018) Law Text Classification Using Semi-Supervised Convolutional Neural Networks. 2018 Chinese Control and Decision Conference (CCDC), Shenyang, 9-11 June 2018, 309-313. [Google Scholar] [CrossRef] 
[5] Ruder, S., Ghaffari, P. and Breslin, J.G. (2016) A Hierarchical Model of Reviews for Aspect-Based Sentiment Analysis. Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing, Austin, 1-5 November 2016, 999-1005. [Google Scholar] [CrossRef] 
[6] Kim, Y. (2014) Convolutional Neural Networks for Sentence Classifi-cation. Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), Doha, 25-29 October 2014, 1746-1751. [Google Scholar] [CrossRef] 
[7] Chen, P., Sun, Z., Bing, L. and Yang, W. (2017) Recurrent Attention Network on Memory for Aspect Sentiment Analysis. Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, Copenhagen, 7-11 September 2017, 452-461. [Google Scholar] [CrossRef] 
[8] Tang, D., Qin, B., Feng, X. and Liu, T. (2015) Effective LSTMs for Target-Dependent Sentiment Classification. ArXiv: 1512.01100.
[9] Bahdanau, D., Cho, K. and Bengio, Y. (2014) Neural Machine Translation by Jointly Learning to Align and Translate. ArXiv: 1409.0473.
[10] Wang, Y., Huang, M., Zhao, L and Zhu, X. (2016) Attention-Based LSTM for Aspect-Level Sentiment Classification. Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing, Austin, 1-5 November 2016, 606-615. [Google Scholar] [CrossRef] 
[11] Ma, D., Li, S., Zhang, X. and Wang, H. (2017) Interactive Attention Networks for Aspect-Level Sentiment Classification. Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence (IJCAI-17), Melbourne, 19-25 August 2017, 4068-4074. [Google Scholar] [CrossRef] 
[12] Wu, Z., Pan, S., Chen, F., et al. (2020) A Comprehensive Survey on Graph Neural Networks. IEEE Transactions on Neural Networks and Learning Systems, 32, 4-24. [Google Scholar] [CrossRef] 
[13] Zhang, C., Li, Q. and Song, D. (2019) Aspect-Based Senti-ment Classification with Aspect-Specific Graph Convolutional Networks. Proceedings of the 2019 Conference on Em-pirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Pro-cessing, Hong Kong, 3-7 November 2019, 4568-4578. [Google Scholar] [CrossRef] 
[14] Wang, B., Guo, P., Wang, X., He, Y. and Wang, W. (2021) Trans-parent Aspect-Level Sentiment Analysis Based on Dependency Syntax Analysis and Its Application on COVID-19. ACM Journal of Data and Information Quality (JDIQ), 14, 1-24. [Google Scholar] [CrossRef] 
[15] Zhang, M. and Qian, T. (2020) Convolution over Hierarchical Syntactic and Lexical Graphs for Aspect Level Sentiment Analysis. Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), Online, 16-20 November 2020, 3540-3549. [Google Scholar] [CrossRef] 
[16] 梁斌, 刘全, 徐进, 等. 基于多注意力卷积神经网络的特定目标情感分析[J]. 计算机研究与发展, 2017, 54(8): 1724-1735.
[17] Pontiki, M., Galanis, D., Pavlopoulos, J., Papageorgiou, H., Androutsopoulos, I. and Manandhar, S. (2014) SemEval-2014 Task 4: Aspect Based Sentiment Analysis. Proceedings of the 8th International Workshop on Semantic Evaluation (SemEval 2014), Dublin, 23-24 August 2014, 27-35. [Google Scholar] [CrossRef] 
[18] Dong, L., Wei, F., Tan, C., et al. (2014) Adaptive Recursive Neural Network for Target-Dependent Twitter Sentiment Classification. Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics, Baltimore, 23-25 June 2014, 49-54. [Google Scholar] [CrossRef] 
[19] Song, Y., Wang, J., Jiang, T., Liu, Z. and Rao, Y. (2019) Attentional Encoder Network for Targeted Sentiment Classification. ArXiv: 1902.09314.
[20] Zhu, X., Zhu, L., Guo, J., Liang, S. and Dietze, S. (2021) GL-GCN: Global and Local Dependency Guided Graph Convolutional Networks for As-pect-Based Sentiment Classification. Expert Systems with Applications, 186, Article ID: 115712. [Google Scholar] [CrossRef]