微信服务号

微信订阅号

2025年6月16日 23:59 星期一
首页 > 过刊浏览>2025年第36卷第6期 >2827-2843. DOI:10.13328/j.cnki.jos.007246
PDF HTML阅读 XML下载 导出引用 引用提醒
基于双语依存关联图的跨语言情感分类
作者:
中图分类号:

TP18

基金项目:

国家自然科学基金(62076175, 61976146); 江苏省双创博士计划


Cross-lingual Sentiment Classification Based on Bilingual Dependency Graph
Author:
  • 摘要
    • | |
  • 访问统计
    • |
  • 参考文献 [60]
    • | | | |
  • 文章评论
    摘要:

    跨语言情感分类在自然语言处理领域非常重要并且已经得到广泛的研究, 因为它可以利用源语言的标签信息构建目标语言的情感分类系统, 从而大大减少目标语言中费时而耗力的标注工作. 不同语言的表达方式存在明显差异是跨语言情感分类的基本挑战, 提出基于双语依存关联图模型的跨语言情感分类方法. 虽然不同语言的表达存在差异, 但是内部的句法依存关系是相似的. 通过在不同语言的词节点之间建边表示双语评论实例的语义相关性, 双语依存关联图能够对不同语言词之间依存关系的相似性进行显式建模, 从而使图神经网络可以在语言内和语言间整合句法结构信息, 进行跨语言情感分类. 利用英文和中文两种语言的数据集进行实验, 实验结果相较于基线方法提高了3%. 研究表明, 利用双语依存关联图能够对不同语言评论实例之间的关联性进行有效建模, 从而显著提升跨语言情感分类的准确率.

    Abstract:

    Cross-lingual sentiment classification is very important in natural language processing and has been widely studied. It uses label information from the source language to construct a sentiment classification system for the target language, thereby greatly reducing the laborious labeling work in the target language. A fundamental challenge in cross-lingual sentiment classification is the obvious difference in the expressions of different languages. This study proposes a method for cross-lingual sentiment classification based on a bilingual dependency graph model. Although the expressions in different languages are various, their internal syntactic dependencies are similar. By establishing edges among word nodes in different languages to represent the semantic relevance of bilingual comment instances, the bilingual dependency graph can explicitly model the similarity of the dependency relationships among words in different languages, allowing graph neural networks to integrate syntactic structure information within and across languages for cross-lingual sentiment classification. Experiments conducted on datasets in both English and Chinese show that the proposed method achieves an improvement of 3% over the baseline method. It is proven that bilingual dependency graphs can effectively model the correlation of comment instances in different languages, thereby significantly improving the accuracy of cross-lingual sentiment classification.

    参考文献
    [1] Chen XL, Sun Y, Athiwaratkun B, Cardie C, Weinberger K. Adversarial deep averaging networks for cross-lingual sentiment classification. Trans. of the Association for Computational Linguistics, 2018, 6: 557–570.
    [2] Ziser Y, Reichart R. Deep pivot-based modeling for cross-language cross-domain transfer with minimal guidance. In: Proc. of the 2018 Conf. on Empirical Methods in Natural Language Processing. Brussels: ACL, 2018. 238–249. [doi: 10.18653/v1/D18-1022]
    [3] Zhang MZ, Fujinuma Y, Boyd-Graber J. Exploiting cross-lingual subword similarities in low-resource document classification. In: Proc. of the 34th AAAI Conf. on Artificial Intelligence. New York: AAAI, 2020. 9547–9554. [doi: 10.1609/aaai.v34i05.6500]
    [4] Devlin J, Chang M W, Lee K, Toutanova K. BERT: Pre-training of deep bidirectional Transformers for language understanding. In: Proc. of the 2019 Conf. of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Vol. 1: Long Papers). Minneapolis: ACL, 2019. 4171–4186. [doi: 10.18653/v1/N19-1423]
    [5] Lample G, Conneau A. Cross-lingual language model pretraining. arXiv:1901.07291, 2019.
    [6] Conneau A, Khandelwal K, Goyal N, Chaudhary V, Wenzek G, Guzmán F, Grave E, Ott M, Zettlemoyer L, Stoyanov V. Unsupervised cross-lingual representation learning at scale. In: Proc. of the 58th Annual Meeting of the Association for Computational Linguistics. ACL, 2020. 8440–8451. [doi: 10.18653/v1/2020.acl-main.747]
    [7] Conneau A, Wu SJ, Li HR, Zettlemoyer L, Stoyanov V. Emerging cross-lingual structure in pretrained language models. In: Proc. of the 58th Annual Meeting of the Association for Computational Linguistics. ACL, 2020. 6022–6034. [doi: 10.18653/v1/2020.acl-main.536]
    [8] Zhao MJ, Schütze H. Discrete and soft prompting for multilingual models. In: Proc. of the 2021 Conf. on Empirical Methods in Natural Language Processing. ACL, 2021. 8547–8555. [doi: 10.18653/v1/2021.emnlp-main.672]
    [9] Lin XV, Mihaylov T, Artetxe M, Wang TL, Chen SH, Simig D, Ott M, Goyal N, Bhosale S, Du JF, Pasunuru R, Shleifer S, Koura PS, Chaudhary V, O’Horo B, Wang J, Zettlemoyer L, Kozareva Z, Diab M, Stoyanov V, Li X. Few-shot learning with multilingual language models. arXiv:2112.10668, 2022.
    [10] Huang LZ, Ma SM, Zhang DD, Wei FR, Wang HF. Zero-shot cross-lingual transfer of prompt-based tuning with a unified multilingual prompt. In: Proc. of the 2022 Conf. on Empirical Methods in Natural Language Processing. Abu Dhabi: ACL, 2022. 11488–11497. [doi: 10.18653/v1/2022.emnlp-main.790]
    [11] Zhang Z, Zhou ZL, Wang YN. SSEGCN: Syntactic and semantic enhanced graph convolutional network for aspect-based sentiment analysis. In: Proc. of the 2022 Conf. of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. Seattle: ACL, 2022. 4916–4925. [doi: 10.18653/v1/2022.naacl-main.362]
    [12] Zhang C, Li QC, Song DW. Aspect-based sentiment classification with aspect-specific graph convolutional networks. In: Proc. of the 2019 Conf. on Empirical Methods in Natural Language Processing and the 9th Int’l Joint Conf. on Natural Language Processing (EMNLP-IJCNLP). Hong Kong: ACL, 2019. 4568–4578. [doi: 10.18653/v1/D19-1464]
    [13] Sun K, Zhang RH, Mensah S, Mao YY, Liu XD. Aspect-level sentiment analysis via convolution over dependency tree. In: Proc. of the 2019 Conf. on Empirical Methods in Natural Language Processing and the 9th Int’l Joint Conf. on Natural Language Processing (EMNLP-IJCNLP). Hong Kong: ACL, 2019. 5679–5688. [doi: 10.18653/v1/D19-1569]
    [14] Birjali M, Kasri M, Beni-Hssane A. A comprehensive survey on sentiment analysis: Approaches, challenges and trends. Knowledge-based Systems, 2021, 226: 107134.
    [15] Zhuang FZ, Qi ZY, Duan KY, Xi DB, Zhu YC, Zhu HS, Xiong H, He Q. A comprehensive survey on transfer learning. Proc. of the IEEE, 2021, 109(1): 43–76.
    [16] Blitzer J, Dredze M, Pereira F. Biographies, Bollywood, boom-boxes and blenders: Domain adaptation for sentiment classification. In: Proc. of the 45th Annual Meeting of the Association of Computational Linguistics. Prague: ACL, 2007. 440–447.
    [17] Yu JF, Jiang J. Learning sentence embeddings with auxiliary tasks for cross-domain sentiment classification. In: Proc. of the 2016 Conf. on Empirical Methods in Natural Language Processing. Austin: ACL, 2016. 236–246. [doi: 10.18653/v1/D16-1023]
    [18] Ziser Y, Reichart R. Pivot based language modeling for improved neural domain adaptation. In: Proc. of the 2018 Conf. of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Vol. 1: Long Papers). New Orleans: ACL, 2018. 1241–1251. [doi: 10.18653/v1/N18-1112]
    [19] Glorot X, Bordes A, Bengio Y. Domain adaptation for large-scale sentiment classification: A deep learning approach. In: Proc. of the 28th Int’l Conf. on Machine Learning. Bellevue: Omnipress, 2011. 513–520.
    [20] Zhou GY, Xie ZW, Huang JX, He TT. Bi-transferring deep neural networks for domain adaptation. In: Proc. of the 54th Annual Meeting of the Association for Computational Linguistics (Vol. 1: Long Papers). Berlin: ACL, 2016. 322–332. [doi: 10.18653/v1/P16-1031]
    [21] Ganin Y, Lempitsky V. Unsupervised domain adaptation by backpropagation. In: Proc. of the 32nd Int’l Conf. on Machine Learning. 2015. 1180–1189.
    [22] Ganin Y, Ustinova E, Ajakan H, Germain P, Larochelle H, Laviolette F, Marchand M, Lempitsky V. Domain-adversarial training of neural networks. The Journal of Machine Learning Research, 2016, 17(1): 2096–2030.
    [23] Li Z, Wei Y, Zhang Y, Yang Q. Hierarchical attention transfer network for cross-domain sentiment classification. In: Proc. of the 32nd AAAI Conf. on Artificial Intelligence. New Orleans: AAAI, 2018. 5852–5859. [doi: 10.1609/aaai.v32i1.12055]
    [24] He RD, Lee WS, Ng HT, Dahlmeier D. Adaptive semi-supervised learning for cross-domain sentiment classification. In: Proc. of the 2018 Conf. on Empirical Methods in Natural Language Processing. Brussels: ACL, 2018. 3467–3476. [doi: 10.18653/v1/D18-1383]
    [25] Ye H, Tan QY, He RD, Li JT, Ng HT, Bing LD. Feature adaptation of pre-trained language models across languages and domains with robust self-training. In: Proc. of the 2020 Conf. on Empirical Methods in Natural Language Processing (EMNLP). ACL, 2020. 7386–7399. [doi: 10.18653/v1/2020.emnlp-main.599]
    [26] Yu JF, Gong CG, Xia R. Cross-domain review generation for aspect-based sentiment analysis. In: Proc. of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th Int’l Joint Conf. on Natural Language Processing (Findings). ACL, 2021. 4767–4777. [doi: 10.18653/v1/2021.findings-acl.421]
    [27] Li JJ, Yu JF, Xia R. Generative cross-domain data augmentation for aspect and opinion co-extraction. In: Proc. of the 2022 Conf. of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. Seattle: ACL, 2022. 4219–4229. [doi: 10.18653/v1/2022.naacl-main.312]
    [28] Yu JF, Zhao QK, Xia R. Cross-domain data augmentation with domain-adaptive language modeling for aspect-based sentiment analysis. In: Proc. of the 61st Annual Meeting of the Association for Computational Linguistics (Vol. 1: Long Papers). Toronto: ACL, 2023. 1456–1470. [doi: 10.18653/v1/2023.acl-long.81]
    [29] Deng Y, Zhang WX, Pan SJ, Bing LD. Bidirectional generative framework for cross-domain aspect-based sentiment analysis. In: Proc. of the 61st Annual Meeting of the Association for Computational Linguistics (Vol. 1: Long Papers). Toronto: ACL, 2023. 12272–12285. [doi: 10.18653/v1/2023.acl-long.686]
    [30] Wan XJ. Co-training for cross-lingual sentiment classification. In: Proc. of the Joint Conf. of the 47th Annual Meeting of the ACL and the 4th Int’l Joint Conf. on Natural Language Processing of the AFNLP. Suntec: ACL, 2009. 235–243.
    [31] Banea C, Mihalcea R, Wiebe J. Multilingual subjectivity: Are more languages better? In: Proc. of the 23rd Int’l Conf. on Computational Linguistics. Beijing: ACL, 2010. 28–36.
    [32] Balamurali AR, Joshi A, Bhattacharyya P. Cross-lingual sentiment analysis for Indian languages using linked WordNets. In: Proc. of the 24th Int’l Conf. on Computational Linguistics (Posters). Mumbai: Indian Institute of Technology Bombay, 2012. 73–82.
    [33] Meng XF, Wei FR, Liu XH, Zhou M, Xu G, Wang HF. Cross-lingual mixture model for sentiment classification. In: Proc. of the 50th Annual Meeting of the Association for Computational Linguistics (Vol. 1: Long Papers). Jeju Island: ACL, 2012. 572–581.
    [34] Lu B, Tan CH, Cardie C, Tsou BK. Joint bilingual sentiment classification with unlabeled parallel corpora. In: Proc. of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies. Portland: ACL, 2011. 320–330.
    [35] Prettenhofer P, Stein B. Cross-language text classification using structural correspondence learning. In: Proc. of the 48th Annual Meeting of the Association for Computational Linguistics. Uppsala: ACL, 2010. 1118–1127.
    [36] Chen Q, Li CL, Li WJ. Modeling language discrepancy for cross-lingual sentiment analysis. In: Proc. of the 2017 ACM on Conf. on Information and Knowledge Management. Singapore: ACM, 2017. 117–126. [doi: 10.1145/3132847.3132915]
    [37] Chandar APS, Lauly S, Larochelle H, et al. An autoencoder approach to learning bilingual word representations. In: Proc. of the 27th Int’l Conf. on Neural Information Processing Systems, Vol. 2. Montreal: MIT Press, 2014. 1853–1861.
    [38] Zhou HW, Chen L, Shi FL, Huang DG. Learning bilingual sentiment word embeddings for cross-language sentiment classification. In: Proc. of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th Int’l Joint Conf. on Natural Language Processing (Vol. 1: Long Papers). Beijing: ACL, 2015. 430–440. [doi: 10.3115/v1/P15-1042]
    [39] Zhou XJ, Wan XJ, Xiao JG. Cross-lingual sentiment classification with bilingual document representation learning. In: Proc. of the 54th Annual Meeting of the Association for Computational Linguistics (Vol. 1: Long Papers). Berlin: ACL, 2016. 1403–1412. [doi: 10.18653/v1/P16-1133]
    [40] Zhou XJ, Wan XJ, Xiao JG. Attention-based LSTM network for cross-lingual sentiment classification. In: Proc. of the 2016 Conf. on Empirical Methods in Natural Language Processing. Austin: ACL, 2016. 247–256. [doi: 10.18653/v1/D16-1024]
    [41] Zou WY, Socher R, Cer D, Manning CD. Bilingual word embeddings for phrase-based machine translation. In: Proc. of the 2013 Conf. on Empirical Methods in Natural Language Processing. Seattle: ACL, 2013. 1393–1398.
    [42] Chen Z, Qian TY. Transfer capsule network for aspect level sentiment classification. In: Proc. of the 57th Annual Meeting of the Association for Computational Linguistics. Florence: ACL, 2019. 547–556. [doi: 10.18653/v1/P19-1052]
    [43] Conneau A, Lample G, Ranzato MA, Denoyer L, Jégou H. Word translation without parallel data. arXiv:1710.04087, 2018.
    [44] Fei HL, Li P. Cross-lingual unsupervised sentiment classification with multi-view transfer learning. In: Proc. of the 58th Annual Meeting of the Association for Computational Linguistics. ACL, 2020. 5759–5771. [doi: 10.18653/v1/2020.acl-main.510]
    [45] Lample G, Conneau A, Denoyer L, Ranzato MA. Unsupervised machine translation using monolingual corpora only. arXiv:1711.00043, 2018.
    [46] Smith SL, Turban DHP, Hamblin S, Hammerla NY. Offline bilingual word vectors, orthogonal transformations and the inverted Softmax. arXiv:1702.03859, 2017.
    [47] Wang K, Shen WZ, Yang YY, Quan XJ, Wang R. Relational graph attention network for aspect-based sentiment analysis. In: Proc. of the 58th Annual Meeting of the Association for Computational Linguistics. ACL, 2020. 3229–3238. [doi: 10.18653/v1/2020.acl-main.295]
    [48] Tang H, Ji DH, Li CL, Zhou QJ. Dependency graph enhanced dual-Transformer structure for aspect-based sentiment classification. In: Proc. of the 58th Annual Meeting of the Association for Computational Linguistics. ACL, 2020. 6578–6588.
    [49] Phan MH, Ogunbona PO. Modelling context and syntactical features for aspect-based sentiment analysis. In: Proc. of the 58th Annual Meeting of the Association for Computational Linguistics. ACL, 2020. 3211–3220. [doi: 10.18653/v1/2020.acl-main.293]
    [50] Huang BX, Carley K. Syntax-aware aspect level sentiment classification with graph attention networks. In: Proc. of the 2019 Conf. on Empirical Methods in Natural Language Processing and the 9th Int’l Joint Conf. on Natural Language Processing (EMNLP-IJCNLP). Hong Kong: ACL, 2019. 5469–5477. [doi: 10.18653/v1/D19-1549]
    [51] Chen CH, Teng ZY, Zhang Y. Inducing target-specific latent structures for aspect sentiment classification. In: Proc. of the 2020 Conf. on Empirical Methods in Natural Language Processing (EMNLP). ACL, 2020. 5596–5607. [doi: 10.18653/v1/2020.emnlp-main.451]
    [52] Liang B, Yin RD, Gui L, Du JC, Xu RF. Jointly learning aspect-focused and inter-aspect relations with graph convolutional networks for aspect sentiment analysis. In: Proc. of the 28th Int’l Conf. on Computational Linguistics. Barcelona: Int’l Committee on Computational Linguistics, 2020. 150–161. [doi: 10.18653/v1/2020.coling-main.13]
    [53] Zhang M, Qian TY. Convolution over hierarchical syntactic and lexical graphs for aspect level sentiment analysis. In: Proc. of the 2020 Conf. on Empirical Methods in Natural Language Processing (EMNLP). ACL, 2020. 3540–3549. [doi: 10.18653/v1/2020.emnlp-main.286]
    [54] Tian YH, Chen GM, Song Y. Aspect-based sentiment analysis with type-aware graph convolutional networks and layer ensemble. In: Proc. of the 2021 Conf. of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. ACL, 2021. 2910–2922. [doi: 10.18653/v1/2021.naacl-main.231]
    [55] Peters ME, Neumann M, Iyyer M, Gardner M, Clark C, Lee K, Zettlemoyer L. Deep contextualized word representations. In: Proc. of the 2018 Conf. of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Vol. 1: Long Papers). New Orleans: ACL, 2018. 2227–2237. [doi: 10.18653/v1/N18-1202]
    [56] Radford A, Narasimhan K, Salimans T, Sutskever I. Improving language understanding with unsupervised learning. Technical Report, OpenAI, 2018. https://openai.com/index/language-unsupervised/
    [57] Wu SJ, Dredze M. Beto, bentz, becas: The surprising cross-lingual effectiveness of BERT. In: Proc. of the 2019 Conf. on Empirical Methods in Natural Language Processing and the 9th Int’l Joint Conf. on Natural Language Processing (EMNLP-IJCNLP). Hong Kong: ACL, 2019. 833–844. [doi: 10.18653/v1/D19-1077]
    [58] Velicković P, Cucurull G, Casanova A, Romero A, Liò P, Bengio Y. Graph attention networks. arXiv:1710.10903, 2018.
    [59] Xu RC, Yang YM. Cross-lingual distillation for text classification. In: Proc. of the 55th Annual Meeting of the Association for Computational Linguistics (Vol. 1: Long Papers). Vancouver: ACL, 2017. 1415–1425. [doi: 10.18653/v1/P17-1130]
    [60] Wang ZY, Liu X, Yang PJ, Liu SX, Wang ZS. Cross-lingual text classification with heterogeneous graph neural network. In: Proc. of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th Int’l Joint Conf. on Natural Language Processing (Vol. 2: Short Papers). ACL, 2021. 612–620. [doi: 10.18653/v1/2021.acl-short.78]
    相似文献
    引证文献
    网友评论
    网友评论
    分享到微博
    发 布
引用本文

白瑞瑞,王中卿,周国栋.基于双语依存关联图的跨语言情感分类.软件学报,2025,36(6):2827-2843

复制
相关视频

分享
文章指标
  • 点击次数:86
  • 下载次数: 779
  • HTML阅读次数: 49
  • 引用次数: 0
历史
  • 收稿日期:2023-11-14
  • 最后修改日期:2024-02-14
  • 在线发布日期: 2024-12-25
文章二维码
友情链接:中国科学院软件研究所中国计算机学会中国科学院国家自然科学基金委员会CCF数字图书馆Int'l J of Softw Info计算机系统应用
您是第20063357位访问者
版权所有：中国科学院软件研究所 京ICP备05046678号-3
地址：北京市海淀区中关村南四街4号,邮政编码：100190
电话：010-62562563 传真：010-62562533 Email：jos@iscas.ac.cn
技术支持：北京勤云科技发展有限公司

京公网安备 11040202500063号