结合共现网络的方面级情感分析研究

doi:10.3778/j.issn.1002-8331.2206-0218

摘要/Abstract

摘要： 方面级情感分类研究中常用的方法是利用句法结构和神经网络的结合模型，但原始句法结构中都未曾考虑共现关系，这会使依赖关系的构建不完整，导致最终效果不佳。提出融合共现网络的图神经网络，重构句法依赖关系和句法依赖标签，根据词汇与方面词之间的距离剪枝，利用关系图注意力网络融合句法依赖关系和文本，通过多头注意力层，将依赖关系与依赖标签相互作用，得到情感特征，在SemEval2014的Restaurant和Laptop数据集以及Twitter数据集上进行实验，该模型的准确率分别达到78.66%、74.19%和73.97%，均高于其余基准模型。

关键词: 情感分析, 关系图注意力网络, 共现网络

Abstract: The common method used in aspect-level emotion classification research is to use a combination model of syntactic structure and neural network, but the co-existing relation is not considered in the original syntactic structure, which makes the construction of dependencies incomplete and leads to poor final results. In this paper, it proposes a graph neural network that fuses the co-occurrence network, first reconstructs syntactic dependencies and syntactic dependency labels, prunes according to the distance between words and aspect words, uses the graph attention network to fuse syntactic dependencies and texts, and then interacts with dependencies with dependency tags through the multi-head attention layer, and finally obtains emotional features, and experiments are conducted on the Restaurant and Laptop in SemEval2014 datasets and Twitter datasets. The accuracy of this model is 78.66%, 74.19% and 73.97%, respectively, which are higher than the remaining benchmark models.

Key words: sentiment analysis, relation graph attention network, co-existing networks

孙天伟, 杨长春, 顾晓清, 谈国胜. 结合共现网络的方面级情感分析研究[J]. 计算机工程与应用, 2023, 59(20): 111-118.

SUN Tianwei, YANG Changchun, GU Xiaoqing, TAN Guosheng. Research on Aspect-Level Sentiment Analysis Combined with Co-Existing Networks[J]. Computer Engineering and Applications, 2023, 59(20): 111-118.

参考文献

[1] 夏鸿斌，李强，刘渊.局部与全局特征融合的方面情感分析网络模型[J].计算机科学与探索，2023，17（4）：902-911.
XIA H B，LI Q，LIU Y.Local and global feature fusion network model for aspect-based sentiment analysis[J].Journal of Frontiers of Computer Science and Technology，2023，17（4）：902-911.
[2] KIRITCHENKO S，ZHU X，CHERRY C，et al.NRC-Canada-2014：detecting aspects and sentiment in customer reviews[C]//Proceedings of the 8th International Workshop on Semantic Evaluation，2014：437-442.
[3] NGUYEN T H，SHIRAI K.PhraseRNN：phrase recursive neural network for aspect-based sentiment analysis[C]//Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing，2015：2509-2514.
[4] BAHDANAU D，CHO K，BENGIO Y.Neural machine translation by jointly learning to align and translate[C]//3rd International Conference on Learning Representations，San Diego，United States，2015.
[5] TANG D，QIN B，LIU T.Aspect level sentiment classification with deep memory network[C]//Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing，2016：214-221.
[6] WANG Y，HUANG M，ZHU X，et al.Attention-based LSTM for aspect-level sentiment classification[C]//Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing，2016：606-615.
[7] MA D，LI S，ZHANG X，et al.Interactive attention networks for aspect-level sentiment classification[C]//Twenty-Sixth International Joint Conference on Artificial Intelligence，2017.
[8] ZHANG C，LI Q C，SONG D W.Aspect-based sentiment classification with aspect-specific graph convolutional networks[C]//Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing（EMNLP-IJCNLP），2019：4568-4578.
[9] 苏锦钿，欧阳志凡，余珊珊.基于依存树及距离注意力的句子属性情感分类[J].计算机研究与发展，2019，56（8）：1731-1745.
SU J D，OUYANG Z F，YU S S.Aspect-level sentiment classification for sentences based on dependency tree and distance attention[J].Journal of Computer Research and Development，2019，56（8）：1731-1745.
[10] RAKHLIN A.Convolutional neural networks for sentence classification[J].arXiv：1408.5882，2014.
[11] XUE W，LI T.Aspect based sentiment analysis with gated convolutional networks[C]//Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics，2018：2514-2523.
[12] 秦旭，杨文忠，王雪颖，等.基于共现关系的多源主题融合模型[J].计算机工程与应用，2020，56（10）：157-162.
QIN X，YANG W Z，WANG X Y，et al.Multi-source topic fusion model based on co-occurrence relation[J].Computer Engineering and Applications，2020，56（10）：157-162.
[13] KIM Y.Convolutional neural networks for sentence classification[C]//Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing，2014：1746-1751.
[14] KALJAHI R，FOSTER J.Any-gram kernels for sentence classification：a sentiment analysis case study[J].arXiv：1712.07004，2017.
[15] KOMBRINK S.Recurrent neural network based language modeling in meeting recognition[C]//Interspeech，2011：2877-2880.
[16] PENG Z，QI Z Y，ZHENG S C，et al.Text classification improved by integrating bidirectional LSTM with two-dimensional max pooling[C]//Proceedings of COLING 2016，the 26th International Conference on Computational Linguistics，2016：3485-3495.
[17] FAN F F，FENG Y S，ZHAO D Y.Multi-grained attention network for aspect-level sentiment classification[C]//Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing，2018：3433-3442.
[18] XU H，LIU B，SHU L，et al.BERT post-training for review reading comprehension and aspect-based sentiment analysis[C]//Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics：Human Language Technologies，2019：2324-2335.
[19] ZHANG M，QIAN T Y.Convolution over hierarchical syntactic and lexical graphs for aspect level sentiment analysis[C]//Empirical Methods in Natural Language Processing，2020：3540-3549.
[20] WANG K，SHEN W Z，YANG Y Y，et al.Relational graph attention network for aspect-based sentiment analysis[C]//Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics，2020：3229-3238.
[21] WANG Q，CHAN A B.Describing like humans：on diversity in image captioning[C]//2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition（CVPR），2020.
[22] 李淑芝，余乐陶，邓小鸿，等.结合Skip-gram和加权损失函数的神经网络推荐模型[J].计算机工程与应用，2020，56（19）：76-85.
LI S Z，YU L T，DENG X H，et al.Neural network recommendation model combined with skip-gram model and weighted loss function[J].Computer Engineering and Applications，2020，56（19）：76-85.
[23] 方炯焜，陈平华，廖文雄.结合GloVe和GRU的文本分类模型[J].计算机工程与应用，2020，56（20）：98-103.
FANG J K，CHEN P H，LIAO W X.Text classification model based on GloVe and GRU[J].Computer Engineering and Applications，2020，56（20）：98-103.
[24] LI D，WEI F R，TAN C Q，et al.Adaptive recursive neural network for target-dependent Twitter sentiment classification[C]//Meeting of the Association for Computational Linguistics，2014：49-54.
[25] 宋威，温子健.基于特征双重蒸馏网络的方面级情感分析[J].中文信息学报，2021，35（7）：126-133.
SONG W，WEN Z J.Feature dual distillation network for aspect-based sentiment analysis[J].Journal of Chinese Information Processing，2021，35（7）：126-133.
[26] WANG W，DAHLMEIER D H R，PAN S J，et al.Aspect-based sentiment analysis：US20180053107A1[P].2018-02-22.
[27] XU K H，ZHAO H，LIU T W.Aspect-specific heterogeneous graph convolutional network for aspect-based sentiment classification[J].IEEE Access，2020，8：139346-139355.