Aspect Level Sentiment Classification Model Based on LaLI and GCN

doi:10.3778/j.issn.1002-8331.2203-0344

Abstract

Abstract: The word lexical and the distance between aspect-words and context are usually ignored in the aspect-level sentiment classification of graph convolution network（GCN） and recurrent neural network, which fails to make the connection and importance between aspect words and their key sentiment ones stand out. To remedy this situation, this paper provides an aspect-level sentiment classification model（LaLI-GCN） with a mixture of LaLI and GCN. Firstly, used bi-directional long short-term memory to extract semantic information and design a fusion algorithm containing the lexical and location information. Secondly, to design an enhancement algorithm in the base of the data from the fusion algorithm above to form a fusion syntactic dependency tree, and GCN is used to fuse the semantic information and syntactic dependency information. Thirdly, it extracts specific aspect features guided by the mask mechanism, and completes emotion classification through interactive attention capture and interactive context. The model experiments in these three public datasets have proved that lexical and location information coming from combined algorithm plays a role in improving emotion discriminatory and is better to conduct sentiment classification than representative models.

Key words: aspect sentiment classification, word lexical, location information, graph convolution network, syntactic dependency tree

摘要： 图卷积网络（graph convolution network，GCN）和循环神经网络的方面级情感分类方法忽略了单词词性信息和方面词与上下文单词之间的位置信息，且难以突出方面与其关键情感词之间的联系和重要性。针对上述问题，提出了一种基于GCN并融合多方面信息的方面情感分类模型（LaLI-GCN）。使用双向长短期记忆网络提取语义信息，并设计融合算法引入单词的词性与位置信息（lexical and location information，LaLI）；结合融合算法的结果用于设计增强算法去生成增强句法依存树，采用GCN融合语义信息和句法依存信息；根据掩码机制提取特定方面特征，利用交互注意力捕捉方面与上下文之间的交互信息完成情感分类。模型在三个公开数据集的实验证明了经过算法融合的词性与位置信息对于提升情感判别的有效性，且相较于当前代表模型有更好的情感分类效果。

关键词: 方面级情感分类, 单词词性, 位置信息, 图卷积网络, 句法依存树

WANG Ze, KONG Weiwei, HUANG Jiyun, ZHANG Mengna, LI Chi. Aspect Level Sentiment Classification Model Based on LaLI and GCN[J]. Computer Engineering and Applications, 2023, 59(13): 110-119.

王泽, 孔韦韦, 黄纪云, 张梦娜, 李驰. 基于LaLI和GCN的方面级情感分类模型[J]. 计算机工程与应用, 2023, 59(13): 110-119.

References

[1] LEI Z，BING L.Sentiment analysis and opinion mining[J].Synthesis Lectures on Human Language Technologies，2016，30（1）：1-167.
[2] ZHOU J，HUANG J X，CHEN Q，et al.Deep learning for aspect?level sentiment classification：survey，vision，and challenges[J].IEEE Access，2019，7：78454-78483.
[3] LIU B，BLASCH E，CHEN Y，et al.Scalable sentiment classification for big data analysis using naive Bayes classifier[C]//Proceedings of 2013 IEEE International Conference on Big Data，October 6-9，2013：99-104.
[4] KAJI N，KITSUREGAWA M.Building lexicon for sentiment analysis from massive collection of HTML documents[C]//Proceedings of the 2007 Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning，June 28-30，2007：1075-1083.
[5] YOUNG T，HAZARIKA D，PORIA S，et al.Recent trends in deep learning based natural language processing[J].IEEE Computational Intelligence Magazine，2018，13（3）：55-75.
[6] LI R，CHEN H，FENG F，et al.Dual graph convolutional networks for aspect-based sentiment analysis[C]//Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing（Volume 1：Long Papers），August 1-6，2021：6319-6329.
[7] CHO K，VAN MERRI?NBOER B，GULCEHRE C，et al.Learning phrase representations using RNN encoder-decoder for statistical machine translation[C]//Proceedings of the Conference on Empirical Methods in Natural Language Processing（EMNLP），October 25-29，2014：1724-1734.
[8] KIM Y.Convolutional neural networks for sentence classification[C]//Proceedings of the Conference on Empirical Methods in Natural Language Processing（EMNLP），October 25-29，2014，1746-1751.
[9] ZHANG M，QIAN T.Convolution over hierarchical syntactic and lexical graphs for aspect level sentiment analysis[C]//Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing（EMNLP），Nov 16-20，2020：3540-3549.
[10] VASWANI A，SHAZEER N，PARMAR N，et al.Attention is all you need[C]//Advances in Neural Information Processing Systems，December 4-9，2017：5998-6008.
[11] HU G，LU G，ZHAO Y.Bidirectional hierarchical attention networks based on document-level context for emotion cause extraction[C]//Findings of the Association for Computational Linguistics（EMNLP 2021），November 7-11，2021：558-568.
[12] TANG D，QIN B，FENG X，et al.Effective LSTMs for target-dependent sentiment classification[C]//Proceedings of the 26th International Conference on Computational Linguistics：Technical Papers，December 11-16，2016：3298-3307.
[13] 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，November 1-5，2016：606-615.
[14] 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，August 19-25，2017：4067-4074.
[15] CHEN P，SUN Z，BING L，et al.Recurrent attention network on memory for aspect sentiment analysis[C]//Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing，September 9-11，2017：452-461.
[16] HUANG B，OU Y，CARLEY K M.Aspect level sentiment classification with attention-over-attention neural networks[C]//Proceedings of the International Conference on Social Computing，Behavioral-Cultural Modeling and Prediction and Behavior Representation in Modeling and Simulation，2018：197-206.
[17] 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，July 15-20，2018：946-956.
[18] CHEN Z，QIAN T.Transfer capsule network for aspect level sentiment classification[C]//Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics，July 28-August 2，2019：547-556.
[19] ZHANG C，LI Q，SONG D.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，November 3-7，2019：4568-4578.
[20] CHEN C，TENG Z，ZHANG Y.Inducing target-specific latent structures for aspect sentiment classification[C]//Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing，Nov 16-20，2020：5596-5607.
[21] 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.
LIANG B，LIU Q，XIU J，et al.Aspect-based sentiment analysis based on multi-attention CNN[J].Journal of Computer Research and Development，2017，54（8）：1724-1735.
[22] SHAFIE A S，SHAREF N M，MURAD M A A，et al.Aspect extraction performance with pos tag pattern of dependency relation in aspect-based sentiment analysis[C]//Proceedings of the 2018 Fourth International Conference on Information Retrieval and Knowledge Management，March 26-28，2018：1-6.
[23] BENGIO Y，DUCHARME R，VINCENT P，et al.A neural probabilistic language model[J].The Journal of Machine Learning Research，2003，3：1137-1155.
[24] KIPF T，WELLING M.Semi-Supervised classification with graph convolutional networks[J].arXiv：1609.02907，2016.
[25] TANG H，JI D，LI C，et al.Dependency graph enhanced dual-transformer structure for aspect-based sentiment classification[C]//Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics，July 5-10，2020：6578-6588