知识增强的交互注意力方面级情感分析模型

doi:10.3778/j.issn.1673-9418.2108082

摘要/Abstract

摘要： 方面级情感分析（ABSA）已经成为自然语言处理领域的研究热点，与传统的情感分析技术相比，基于方面的情感分析能够判断句子中多个方面的情感倾向，可以更加准确地挖掘用户对方面的情感极性。当前，将注意力机制和神经网络相结合的模型在解决方面级情感分析任务时大多仅考虑方面对上下文的影响，且时常忽略句子中的相关语法信息和背景知识。针对上述问题，提出一种借助知识图谱和图卷积网络的交互注意力神经网络模型，为评论文本注入背景信息和语言知识。首先，利用知识图谱解决词汇在不同语境下的一词多义性问题。其次，利用文本图卷积网络完善评论语句的语法结构信息。最后，通过交互注意力机制实现评论文本上下文与评价方面的协调优化。最终在五个公开数据集上的实验结果表明，合理利用外部知识是改善方面级情感分析模型性能的有效策略。

关键词: 知识图谱, 词汇句法关系, 图神经网络, 方面级情感分析, 交互注意力机制

Abstract: Aspect-based sentiment analysis (ABSA) has become one of the hottest research issues in the field of natural language processing. Compared with traditional sentiment analysis technology, aspect-based sentiment analy-sis can judge the sentiment tendency of multiple targets in a sentence, and more accurately mine the sentiment polarity of the aspect. Currently, the models combining attention mechanism and neural network only consider the impact of aspects on the context, and often ignore the context information of sentences and background knowledge. To solve the above problems, an interactive attention neural network model based on knowledge maps and graph convolution network is proposed to inject background information and language knowledge into review text. Firstly, the polysemy problem of vocabulary under different contexts is addressed via knowledge maps. Secondly, text graph convolution networks are used to improve the syntactic structure information of review text. Finally, the context and aspect of the review text are coordinated and optimized through an interactive attention mechanism. Experimental results on five public datasets show that rational use of external knowledge is an effective strategy for enhancing the performance of aspect-based sentiment analysis model.

Key words: knowledge graph, lexical and syntactic relations, graph neural networks, aspect-based sentiment analysis, interactive attention mechanism

韩虎, 郝俊, 张千锟, 孟甜甜. 知识增强的交互注意力方面级情感分析模型[J]. 计算机科学与探索, 2023, 17(3): 709-718.

HAN Hu, HAO Jun, ZHANG Qiankun, MENG Tiantian. Knowledge-Enhanced Interactive Attention Model for Aspect-Based Sentiment Analysis[J]. Journal of Frontiers of Computer Science and Technology, 2023, 17(3): 709-718.

参考文献

[1] MEDHATW, HASSAN A, KORASHY H. Sentiment analy-sis algorithms and applications: a survey[J]. Ain Shams Engineering Journal, 2014, 5(4): 1093-1113.
[2] PONTIKI M, GALANIS D, PAPAGEORGIOU H, et al. Sem-Eval-2014 Task 4: aspect based sentiment analysis[C]//Pro-ceedings of the 8th International Workshop on Semantic Evaluation, Dublin, Aug 23-24, 2014. Stroudsburg: ACL, 2014:?27-35.
[3] BOIY E, MOENS M F. A machine learning approach to sen-timent analysis in multilingual Web texts[J]. Information Retrieval, 2009, 12(5): 526-558.
[4] KIM Y. Convolutional neural networks for sentence classifi-cation[C]//Proceedings of the 2014 Conference on Empiri-cal Methods in Natural Language Processing, Doha, Oct 5-29, 2014. Stroudsburg: ACL, 2014: 1746-1751.
[5] WANG S, MAZUMDER S, LIU B, et al. Target-sensitive me-mory networks for aspect sentiment classification[C]//Pro-ceedings of the 56th Annual Meeting of the Association for Computational Linguistics, Melbourne, Jul 15-20, 2018. Stroud-sburg: ACL, 2018: 957-967.
[6] MA D H, LI S J, ZHANG X D, et al. Interactive attention net-works for aspect-level sentiment classification[C]//Procee-dings of the 26th International Joint Conference on Artifi-cial Intelligence, Melbourne, Aug 19-25, 2017: 4068-4074.
[7] TAY Y, TUAN L A, HUI S C. Learning to attend via word-aspect associative fusion for aspect-based sentiment analy-sis[C]//Proceedings of the 2018 AAAI Conference on Artifi-cial Intelligence, New Orleans, Feb 2-7, 2018. Palo Alto: AAAI, 2018: 5956-5963.
[8] YAO L, MAO C S, LUO Y. Graph convolutional networks for text classification[C]//Proceedings of the 33rd AAAI Confe-rence on Artificial Intelligence, the 31st Innovative Applica-tions of Artificial Intelligence Conference, the 9th AAAI Sympo-sium on Educational Advances in Artificial Intelligence, Hono-lulu, Jan 27-Feb 1, 2019. Menlo Park: AAAI, 2019: 7370-7377.
[9] SHAFIE A S, SHAREF N M, MURAD M A A, et al. Aspect extraction performance with POS tag pattern of depen-dency relation in aspect-based sentiment analysis[C]//Pro-ceedings of the 2018 4th International Conference on Infor-mation Retrieval and Knowledge Management, Kota Kina-balu, Mar 26-28, 2018.?Piscataway: IEEE, 2018: 107-112.
[10] ZHANG C, LI Q C, SONG D W. Aspect-based sentiment classification with aspect-specific graph convolutional net-works[C]//Proceedings of the 2019 Conference on Empiri-cal Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Proces-sing, Hong Kong, China, Nov 3-7, 2019. Stroudsburg: ACL, 2019: 4567-4577.
[11] CHEN J D, HU Y Z, LIU J P, et al. Deep short text clas-sification with knowledge powered attention[C]//Procee-dings of the 33rd AAAI Conference on Artificial Intel-ligence, the 31st Innovative Applications of Artificial Intelli-gence Conference, the 9th AAAI Symposium on Educa-tional Advances in Artificial Intelligence, Honolulu, Jan 27-Feb 1, 2019. Menlo Park: AAAI, 2019: 6252-6259.
[12] BIAN X O, FENG C, AHMAD A, et al. Targeted sentiment classification with knowledge powered attention network[C]//Proceedings of the 31st IEEE International Conference on Tools with Artificial Intelligence, Portland, Nov 4-6, 2019. Piscataway: IEEE, 2019: 1073-1080.
[13] TAI K S, SOCHER R, MANNING C D. Improved seman-tic representations from tree-structured long short-term me-mory networks[C]//Proceedings of the 53rd Annual Mee-ting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing of the Asian Federation of Natural Language Processing, Beijing, Jul 26-31, 2015. Stroudsburg: ACL, 2015: 1556-1566.
[14] YANG M, TU W, WANG J, et al. Attention-based LSTM for target-dependent sentiment classification[C]//Proceedings of the 31st AAAI Conference on Artificial Intelligence, San Francisco, Feb 4-9, 2017. Menlo Park: AAAI, 2017: 5013-5014.
[15] TANG D Y, QIN B, LIU T. Aspect level sentiment clas-sification with deep memory network[C]//Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing, Austin, Nov 1-4, 2016. Stroudsburg: ACL, 2016: 214-224.
[16] SUN K, ZHANG R, MENSAH S, et al. Aspect-level senti-ment analysis via convolution over dependency tree[C]//Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing, Hong Kong, China, Nov 3-7, 2019. Stroudsburg: ACL, 2019: 5683-5692.
[17] ZHANG M, QIAN T. Convolution over hierarchical syntac-tic and lexical graphs for aspect level sentiment analysis[C]//Proceedings of the 2020 Conference on Empirical Me-thods in Natural Language Processing. Stroudsburg: ACL, 2020: 3540-3549.
[18] TIAN Y H, CHEN G M, SONG Y. Enhancing aspect-level sentiment analysis with word dependencies[C]//Procee-dings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Vo-lume. Stroudsburg: ACL, 2021: 3726-3739.
[19] HU L M, YANG T C, SHI C, et al. Heterogeneous graph attention networks for semi-supervised short text classifica-tion[C]//Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Proces-sing, Hong Kong, China, Nov 3-7, 2019. Stroudsburg: ACL, 2019: 4820-4829.
[20] JI L, WANG Y J, SHI B T, et al. Microsoft concept graph: mining semantic concepts for short text understanding[J]. Data Intelligence, 2019, 1(3): 238-270.
[21] SUCHANEK F M, KASNECI G, WEIKUM G. YAGO: a large ontology from Wikipedia and WordNet[J]. Journal of Web Semantics, 2008, 6(3): 203-217.
[22] MILLER G A. WordNet: a lexical database for English[J]. Communications of the ACM, 1995, 38(11): 39-41.
[23] DONG L, WEI F R, TAN C Q, et al. Adaptive recursive neural network for target-dependent Twitter sentiment clas-sification[C]//Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics, Baltimore, Jun 23-25, 2014. Stroudsburg: ACL, 2014: 49-54.
[24] 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, Denver, Jun 4-5, 2015. Stroudsburg: ACL, 2015: 486-495.
[25] PONTIKI M, GALANIS D, PAPAGEORGIOU H, et al.SemEval-2016 Task 5: aspect based sentiment analysis[C]//Proceedings of the 10th International Workshop on Se-mantic, Evaluation, San Diego, Jun 16-17, 2016. Stroudsburg: ACL, 2016: 19-30.
[26] PENNINGTON J, SOCHER R, MANNING C D. GloVe: global vectors for word representation[C]//Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing, Doha, Oct 25-29, 2014. Stroudsburg: ACL, 2014: 1532-1543.
[27] 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 Lin-guistics: Technical Papers, Osaka, Dec 11-17, 2016. Strouds-burg: ACL, 2016: 3298-3307.