基于多定序尺度的图卷积网络方面级情感分析

doi:10.3778/j.issn.1002-8331.2404-0247

摘要/Abstract

摘要： 方面级情感分析旨在预测给定句子中特定方面的情感极性。由于现有研究方法仅考虑单一语料信息，导致存在语料库词共现、评论语句语义结构、局部词汇序列等多种信息利用不全面和不充分的问题，为此提出一种基于多定序尺度的图卷积网络模型。定序尺度分为语料库级、句子级、词汇级。构造包含三个卷积模块的多定序尺度图卷积网络：纵向全局语法依赖图卷积引入语料库全局词共现信息丰富语法表示；横向全局语义结构图卷积学习句子本身的语义结构信息；文本局部序列图卷积挖掘句子局部范围内的词汇序列信息。利用混合注意力进一步增强语法和语义结构特征表示，并通过交互机制实现多种信息的共享与融合。实验结果表明，该模型与经典模型ASGCN相比，在五个公开数据集上准确率分别提升了2.56、2.93、3.13、2.58、1.76个百分点。与最新模型DGGCN和CRF-GCN相比，均取得了优异的情感分类效果，证实了该模型融合多种语料信息的有效性。

关键词: 方面级情感分析, 图卷积网络, 多定序尺度, 混合注意力

Abstract: Aspect-based sentiment analysis aims to predict the sentiment polarity of specific aspects in a given sentence. Since the existing research methods only consider single corpus information, which leads to the problem of incomplete and insufficient utilization of multiple information such as corpus word co-occurrence, semantic structure of commented utterances, and local lexical sequences, a graph convolutional network model based on multiple ordinal scales is proposed for this purpose. The ordinal scales are categorized into corpus level, sentence level, and lexical level. Firstly, a multiple ordinal scale graph convolutional network containing three convolutional modules is constructed, in which vertical global syntactic dependency graph convolution introduces the corpus global word co-occurrence information to enrich the syntactic representation; horizontal global semantic structure graph convolution learns the semantic structure information of the sentence itself; and textual local sequence graph convolution mines the vocabulary sequence information of the sentence in the local scope. Secondly, hybrid attention is utilized to further enhance the syntactic and semantic structural feature representations, and the sharing and fusion of multiple information is achieved through the interaction mechanism. Finally, the experimental results show that the model improves the accuracy by 2.56, 2.93, 3.13, 2.58, and 1.76 percentage points on five public datasets compared with the classical model ASGCN. Compared with the latest models DGGCN and CRF-GCN, both achieve excellent sentiment classification results, confirming the effectiveness of the model in fusing multiple corpus information.

Key words: aspect-based sentiment analysis, graph convolutional network, multiple ordinal scales, hybrid attention

穆一茹, 韩虎, 孔博. 基于多定序尺度的图卷积网络方面级情感分析[J]. 计算机工程与应用, 2025, 61(14): 238-247.

MU Yiru, HAN Hu, KONG Bo. Aspect-Based Sentiment Analysis Based on Multiple Ordinal Scales of Graph Convolutional Network[J]. Computer Engineering and Applications, 2025, 61(14): 238-247.

参考文献

[1] TANG D, QIN B, FENG X, et al. Target-dependent sentiment classification with long short term memory[C]//Proceedings of the 26th International Conference on Computational Linguistics. Stroudsburg: ACL, 2016: 3298-3307.
[2] WANG Y, HUANG M, ZHU X, et al. Attention-based LSTM for aspect?level sentiment classification[C]//Proceedings of the Conference on Empirical Methods Innatural Language Processing. Stroudsburg: ACL, 2016: 606-615.
[3] FAN F, FENG Y, ZHAO D. Multi-grained attention network for aspect-level sentiment classification[C]//Proceedings of the Conference on Empirical Methods in Natural Language Processing. Stroudsburg: ACL, 2018: 3433-3442.
[4] 袁勋, 刘蓉, 刘明. 融合多层注意力的方面级情感分析模型[J]. 计算机工程与应用, 2021, 57(22): 147-152.
YUAN X, LIU R, LIU M. Aspect-level sentiment analysis model incorporating multi-layer attention[J]. Computer Engineering and Applications, 2021, 57(22): 147-152.
[5] HUANG B, 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. Stroudsburg: ACL, 2019: 5469-5477.
[6] ZHANG C, LI Q, SONG D. Aspect-based sentiment classification with aspect-specific graph convolutional networks[C]//Proceedings of the Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing. Stroudsburg: ACL, 2019: 4568-4578.
[7] WANG K, SHEN W, YANG 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, Online. Stroudsburg: ACL, 2020: 3229-3238.
[8] BAI X, LIU P, ZHANG Y. Investigating typed syntactic dependencies for targeted sentiment classification using graph attention neural network[J]. IEEE/ACM Transactions on Audio, Speech, and Language Processing, 2020, 29: 503-514.
[9] 王汝言, 陶中原, 赵容剑, 等. 多交互图卷积网络用于方面情感分析[J]. 电子与信息学报, 2022, 44(3): 1111-1118.
WANG R Y, TAO Z Y, ZHAO R J et al. Multi-interaction graph convolutional networks for aspect-level sentiment analysis[J]. Journal of Electronics and Information Technology, 2022, 44(3): 1111-1118.
[10] YAO L, MAO C, LUO Y. Graph convolutional networks for text classification[C]//Proceedings of the 33rd AAAI Conference on Artificial Intelligence and 31st Innovative Applications of Artificial Intelligence Conference and Ninth AAAI Symposium on Educational Advances in Artificial Intelligence. Washington DC: AAAI, 2019: 7370-7377.
[11] PENNINGTON J, SOCHER R, MANNING C. GloVe: global vectors for word representation[C]//Proceedings of the Conference on Empirical Methods in Natural Language Processing. Stroudsburg: ACL, 2014: 1532-1543.
[12] DEVLIN J, CHANG M W, LEE K, et al. BERT: pretraining of deep bidirectional transformers for language understanding[C]//Proceedings of the Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. Stroudsburg: ACL, 2019: 4171-4186.
[13] OEPEN S, KUHLMANN M, MIYAO Y, et al. SemEval 2015 task 18: broad-coverage semantic dependency paing[C]//Proceedings of the 9th International Workshop on Semantic Evaluation. Stroudsburg: ACL, 2015: 915-926.
[14] JIANG Z H, YU W, ZHOU D, et al. ConvBERT: improving BERT with spanbased dynamic convolution[C]//Advances in Neural Information Processing Systems, 2020: 12837-12848.
[15] WU F, FAN A, BAEVSKI A, et al. Pay less attention with lightweight and dynamic convolutions[J]. arXiv:1901.10430, 2019.
[16] LI D, WEI F, TAN C, et al. Adaptive recursive neural network for target dependent twitter sentiment classification[C]//Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics. Stroudsburg: ACL, 2014: 49-54.
[17] PONTIKI M, GALANIS D, PAVLOPOULOS J, et al. Semeval-2014 task 4: aspect based sentiment analysis[C]//Proceedings of the 8th International Workshop on Semantic Evaluation. Stroudsburg: ACL, 2014: 27-35.
[18] PONTIKI M, GALANIS D, PAPAGEORGIOU H, et al. Semeval-2015 task 12: aspect based sentiment analysis[C]//Proceedings of the 9th International Workshop on Semantic Evaluation. Stroudsburg: ACL, 2015: 486-495.
[19] PONTIKI M, GALANIS D, PAPAGEORGIOU H, et al. Semeval-2016 task 5: aspect based sentiment analysis[C]//Proceedings of the 10th International Workshop on Semantic Evaluation. Stroudsburg: ACL, 2016: 19-30.
[20] KIM Y. Convolutional neural networks for sentence classification[C]//Proceedings of the Conference on Empirical Methods in Natural Language Processing. Stroudsburg: ACL, 2014: 1746-1751.
[21] TAY Y, TUAN L A, HUI S C. Learning to attend via wordaspect associative fusion for aspect-based sentiment analysis[C]//Proceedings of the 32nd AAAI Conference on Artificial Intelligence. Washington DC: AAAI, 2018: 5956-5963.
[22] ZHAO M, YANG J, ZHANG J, et al. Aggregated graph convolutional networks for aspect-based sentiment classification[J]. Information Sciences, 2022, 600: 73-93.
[23] ZHOU J, HUANG J X, HU Q V, et al. SK-GCN: modeling syntax and knowledge via graph convolutional network for aspect-level sentiment classification[J]. Knowledge-Based Systems, 2020, 205: 106292.
[24] WANG X, LIU P, ZHU Z, et al. Aspect-based sentiment analysis with graph convolutional networks over dependency awareness[C]//Proceedings of the 26th International Conference on Pattern Recognition. Piscataway: IEEE, 2022: 2238-2245.
[25] LIU H, WU Y, LI Q, et al. Enhancing aspect-based sentiment analysis using a dual-gated graph convolutional network via contextual affective knowledge[J]. Neurocomputing, 2023, 553: 126526.
[26] HUANG B, ZHANG J, JU J, et al. CRF-GCN: an effective syntactic dependency model for aspect-level sentiment analysis[J]. Knowledge-Based Systems, 2023, 260: 110125.
[27] 韩虎, 孔博, 何勇禧, 等. 基于剪枝策略的知识增强方面级情感分析[J]. 华中科技大学学报 (自然科学版), 2024, 52(11): 140-146.
HAN H, KONG B, HE Y X, et al. Aspect based sentiment analysis based on knowledge enhancement of pruning strategy[J]. Journal of Hua Zhong University of Science and Technology (Natural Science Edition), 2024, 52(11): 140-146.