方面级情感分析作为情感分析的一项细粒度任务,具有非常高的研究价值。方面词和对应的情感词之间的联系对于确定情感极性起着至关重要的作用。先前的研究大多仅利用一种注意力机制来关注句子和目标之间的联系,未考虑到词性中包含的情感信息。为解决这一问题,该文提出了一种基于ELMo的混合注意力网络(ELMo-based Hybrid Attention Network, EHAN)。与现有网络不同的是,模型不仅将ELMo与Transformer网络相结合来捕获文本信息的情感特征,还利用词性注意力机制对词性和单词进行交互获得方面与情感词之间的联系。在公开数据集上的实验结果表明,EHAN与基准模型相比在准确率和Macro-F1值上都有显著提升,证明该方法可有效改善方面级情感分析的性能。
Abstract
As a fine-grained task of sentiment analysis, aspect-level sentiment analysis has very high research value. The interaction between aspect words and corresponding emotional words plays a vital role in determining emotional polarity. In contrast to the existing studies with an attention mechanism to focus on the connection between the sentence and the target, this paper proposes a hybrid attention network (EHAN) based on ELMo. Specifically, the model combines ELMo and Transformer network to capture the emotional characteristics of text information, introducing the an additional mechanism to interact with parts-of-speech and words. The experimental results show that EHAN achieves a significant improvement in accuracy and Macro-F1 compared with the benchmark model.
关键词
方面级情感分析 /
文本表示 /
词性信息 /
注意力机制
{{custom_keyword}} /
Key words
aspect-level sentiment analysis /
text representation /
part-of-speech information /
attention mechanism
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] LIU B. Sentiment analysis and opinion mining[M].Morgan& Claypool Publishers, 2012: 1-167.
[2] LI X, LAM W. Deep multi-task learning for aspect term extraction with memory interaction[C]//Proceedings of the Conference on Empirical Methods in Natural Language Processing, 2017: 2886-2892.
[3] WANG W Y, PAN S J, DAHLMEIER D, et al. Coupled multi-layer attentions for co-extraction of aspect and opinion terms[C]//Proceedings of the 31st AAAI Conference on Artificial Intelligence, 2017: 3316-3322.
[4] TANG D Y, QIN B, FENG X C, et al. Effective LSTMS for target-dependent sentiment classification[C]//Proceedings of the 26th International Conference on Computational Linguistics, 2016: 3298-3307.
[5] WANG X Y, XU G L, ZHANG J Y, et al. Syntax-directed hybrid attention network for aspect-level sentiment analysis[J]. IEEE Access, 2019: 5014-5025.
[6] 杜成玉,刘鹏远.基于螺旋注意力网络的方面级别情感分析模型[J].中文信息学报,2020,34(9): 70-77.
[7] ZHANG S Q, XU X Y, PANG Y W, et al. Multi-layer attention based CNN for target-dependent sentiment classification[J]. Neural Processing Letters,2020,51(3): 2089-2103.
[8] 曾锋,曾碧卿,韩旭丽,等.基于双层注意力循环神经网络的方面级情感分析[J].中文信息学报,2019,33(06): 108-115.
[9] 王行甫,王磊,苗付友,等.结合词性、位置和单词情感的内存网络的方面情感分析[J].小型微型计算机系统,2019,40(02): 383-389.
[10] PETERS M E, NEUMANN M,IYYER M, et al. Deep contextualized word representations[C]//Proceedings of the Conference of the North American Chapter of the Association for Computational Linguistics,2018: 2227-2237
[11] VASWANI A,SHAZEER N, PARMAR N, et al. Attention is all you need[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems,2017: 6000-6010.
[12] HUANG B X,OU Y L, CARLEY K M. Aspect level sentiment classification with attention-over-attention neural networks[C]//Proceedings of International Conference on Social Computing,2018: 197-206.
[13] 赵亚欧,张家重,李贻斌,等.融合基于语言模型的词嵌入和多尺度卷积神经网络的情感分析[J].计算机应用,2020,40(03): 651-657.
[14] KATARZYNA B,BIESIALSKA M, HENRYK R. Sentiment analysis with contextual embeddings and self-attention[C]//Proceedings of International Syposium on Methodologies for Intelligent Systems,2020: 32-41.
[15] 王家乾,龚子寒,薛云,等.基于混合多头注意力和胶囊网络的特定目标情感分析[J].中文信息学报,2020,34(05): 100-110.
[16] XU Q N, ZHU L, DAI T, et al. Aspect-based sentiment classification with multi-attention network[J]. Neurocomputing, 2020: 135-143.
[17] ZHANG X S, GAO T. Multi-head attention model for aspect level sentiment analysis[J]. Journal of Intelligent and Fuzzy Systems, 2020, 38(1): 89-96.
[18] HUANG B X, CARLEY K M. Syntax-aware aspect level sentiment classification with graph attention networks[C]//Proceedings of the Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing, 2019: 5468-5476.
[19] CHENG K F, YUE Y A, SONG Z W. Sentiment Classification based on part-of-speech and self-attention mechanism[J]. IEEE Access, 2020,8: 16387-16396.
[20] 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.
[21] MA D, LI S, ZHANG X, et al. Interactive attention networks for aspect-level sentiment classification[C]//Proceedings of the 26th International Joint Conference on Artificial Intelligence, 2017: 4068-4074.
[22] GU S Q, ZHANG L P, HOU Y X, et al. A position-aware bidirectional attention network for aspect-level sentiment analysis[C]//Proceedings of the 27th International Conference on Computational Linguistics, 2018: 774-784.
[23] LI X, BING L, LAM W, et al. Transformation networks for target-oriented sentiment classification[C]//Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics, 2018: 946-956.
[24] SONG Y W, WANG J H, JIANG T, et al. Attentional encoder network for targeted sentiment classification[J].arXiv preprint arXiv: 1902.09314v2.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
PDF(3914 KB)
