复合双通道酒店评论方面类别情感分析
Composite Dual-Channel Sentiment Analysis of Hotel Reviews in Terms of Categories
摘要: 针对中文酒店评论情感分析会忽视特定方面情感极性的问题，提出一种复合双通道模型GCGAT，进行方面类别情感分析。该模型基于门控机制和双向门控循环单元，由四层网络结构组成；利用门控机制控制通道输出；引入方面类别影响因素词向量和注意力机制提高模型信息提取准确性；数据集方面，利用LDA-Jaccard模型进行聚类，确定方面类别并进行标注，建立了细粒度数据集。通过对比实验发现，GCGAT模型取得了很好的效果，在准确率、F1值、ROC曲线和auc值上效果均优于其他模型。
Abstract: Aiming at the problem that the sentiment analysis of Chinese hotel reviews will ignore the polarity of specific aspectual sentiment, a composite two-channel model, GCGAT, is proposed for aspectual category sentiment analysis. The model is based on the gating mechanism and bidirectional gating loop unit, and consists of a four-layer network structure; the gating mechanism is utilized to control the channel output; the aspect category influencing factor word vector and the attention mecha-nism are introduced to improve the accuracy of the model’s information extraction; in terms of the dataset, clustering is carried out by using the LDA-Jaccard model to determine the aspect categories and annotate them, and a fine-grained dataset is established. Through comparison experiments, it was found that the GCGAT model achieved good results, and the effect was better than other models in terms of accuracy, F1 value, ROC curve and auc value.
文章引用：李薇, 王欣羽. 复合双通道酒店评论方面类别情感分析[J]. 计算机科学与应用, 2023, 13(11): 2000-2011. https://doi.org/10.12677/CSA.2023.1311198

参考文献

[1] Wang, J., Du, J.P., Shao, Y.X. and Li, A. (2022) Sentiment Analysis of Online Travel Reviews Based on Capsule Net-work and Sentiment Lexicon. In: Zhang, L., Yu, W., Jiang, H. and Laili, Y., Eds., CINT 2022: Intelligent Networked Things, Springer, Singapore, 92-106.
[2] 邓慈云, 余国清. 基于朴素贝叶斯的影评情感分析研究[J]. 智能计算机与应用, 2023, 13(2): 210-212, 217.
[3] Tong, X., Li, Y.X., Li, J.Y., et al. (2022) What Are People Talking about in #BlackLivesMatter and #StopAsianHate? Exploring and Categorizing Twitter Topics Emerging in Online Social Move-ments through the Latent Dirichlet Allocation Model. Proceedings of the 2022 AAAI/ACM Conference on AI, Ethics, and Society, Oxford, 19-21 May 2021, 723-738. [Google Scholar] [CrossRef] 
[4] 孔繁钰, 陈纲. 基于改进双向LSTM的评教文本情感分析[J]. 计算机工程与设计, 2022, 43(12): 3580-3587.
[5] Wei, J.Y., Liao, J., Yang, Z.F., et al. (2020) BiLSTM with Multi-Polarity Orthogonal Attention for Implicit Sentiment Analysis. Neuro-computing, 383, 165-173. [Google Scholar] [CrossRef] 
[6] 刘继, 顾凤云. 基于BERT与BiLSTM混合方法的网络舆情非平衡文本情感分析[J]. 情报杂志, 2022, 41(4): 104-110.
[7] Phan, H.T., Nguyen, N.T. and Hwang, D. (2022) Aspect-Level Sentiment Analysis Using CNN over BERT-GCN. IEEE Access, 10, 110402-110409. [Google Scholar] [CrossRef] 
[8] 曹卫东, 潘红坤. 采用稀疏自注意力机制和BiLSTM模型的细粒度情感分析[J]. 计算机应用与软件, 2022, 39(12): 187-194.
[9] Yang, J. and Yang, J. (2020) Aspect Based Sentiment Analysis with Self-Attention and Gated Convolutional Networks. 2020 IEEE 11th In-ternational Conference on Software Engineering and Service Science, Beijing, 16-18 October 2020, 146-149. [Google Scholar] [CrossRef] 
[10] Fei, H., Li, F., Li, C.L., et al. (2022) Inheriting the Wisdom of Predecessors: A Multiplex Cascade Framework for Unified Aspect-Based Sentiment Analysis. Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence Main Track, Vienna, 23-29 July 2022, 4121-4128. [Google Scholar] [CrossRef] 
[11] 中华人民共和国国家统计局. 中国统计年鉴[M]. 北京: 中国统计出版社, 2022.
[12] 邬小燕. 基于Stacking的酒店评论情感分析研究[D]: [硕士学位论文]. 重庆: 重庆大学, 2019.
[13] 高华玲, 张晶. 基于情感词典的酒店评论情感分析与可视化[J]. 软件, 2021, 42(1): 45-47, 66.
[14] 丁美荣, 冯伟森, 黄荣翔, 等. 基于预训练模型和基础词典扩展的酒店评论情感分析[J]. 计算机系统应用, 2022, 31(11): 296-308.
[15] Xue, W. and Li, T. (2018) Aspect-Based Sentiment Analysis with Gated Convolutional Networks. Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics, Melbourne, July 2018, 2514-2523. [Google Scholar] [CrossRef] 
[16] Vaswani, A., Shazeer, N., Parmar, N., et al. (2017) Attention Is All You Need. arXiv: 1706.03762.
[17] Kiritchenko, S., Zhu, X.D., Cherry, C. and Mohammad, S. (2014) NRC-Canada-2014: Detecting Aspects and Sentiment in Customer Reviews. Proceedings of the 8th International Work-shop on Semantic Evaluation (SemEval 2014), Dublin, August 2014, 437-442. [Google Scholar] [CrossRef] 
[18] 王新宇. 基于情感词典与机器学习的旅游网络评价情感分析研究[J]. 计算机与数字工程, 2016, 44(4): 578-582, 766.
[19] Mikolov, T., Sutskever, I., Chen, K., et al. (2013) Distributed Representations of Words and Phrases and Their Compositionality. arXiv: 1310.4546.
[20] 徐菲菲, 芦霄鹏. 结合卷积神经网络和最小门控单元注意力的文本情感分析[J]. 计算机应用与软件, 2020, 37(9): 75-80, 125.
[21] Tang, F.L., Fu, L.Y., Yao, B., et al. (2019) Aspect Based Fine-Grained Sentiment Analysis for Online Reviews. Information Sciences, 488, 190-204. [Google Scholar] [CrossRef] 
[22] Tang, D.Y., Qin, B., Feng, X.C., et al. (2016) Effective LSTMs for Target-Dependent Sentiment Classification. Proceedings of COLING 2016, the 26th Inter-national Conference on Computational Linguistics: Technical Papers, Osaka, December 2016, 3298-3307.
[23] Xue, W. and Li, T. (2018) Aspect-Based Sentiment Analysis with Gated Convolutional Networks. Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics, Melbourne, July 2018, 2514-2523. [Google Scholar] [CrossRef] 
[24] Yao, L.S. (2022) Sentiment Analysis Based on CNN-LSTM Hotel Re-views. Journal of Physics: Conference Series, 2330, Article ID: 012018. [Google Scholar] [CrossRef] 
[25] 马桂真, 彭霞. 基于酒店评论大数据的游客评论主题挖掘与情感分析——以北京五星级酒店为例[J]. 北京联合大学学报, 2021, 35(2): 58-68.
[26] 王宏鹏. 基于词典与机器学习的酒店评论情感分析的研究[D]: [硕士学位论文]. 大连: 大连交通大学, 2020.
[27] 池毛毛, 潘美钰, 王伟军. 共享住宿与酒店用户评论文本的跨平台比较研究: 基于LDA的主题社会网络和情感分析[J]. 图书情报工作, 2021, 65(2): 107-116.
[28] Dauphin, Y.N., Fan, A., Auli, M. and Grangier, D. (2017) Language Modeling with Gated Convolutional Networks. Proceedings of the 34th International Conference on Machine Learning, 70, 933-941.
[29] Jonathan, T., Ross, G., Arjun, J., et al. (2015) Efficient Object Localization Using Convolutional Net-works. 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Boston, 7-12 June 2015, 648-656.