隐喻是人类语言中经常出现的一种特殊现象,隐喻识别对于自然语言处理各项任务来说具有十分基础和重要的意义。针对中文领域的隐喻识别任务,该文提出了一种基于句法感知图卷积神经网络和ELECTRA的隐喻识别模型(Syntax-aware GCN with ELECTRA, SaGE)。该模型从语言学出发,使用ELECTRA和Transformer编码器抽取句子的语义特征,将句子按照依存关系组织成一张图并使用图卷积神经网络抽取其句法特征,在此基础上对两类特征进行融合以进行隐喻识别。该模型在CCL 2018中文隐喻识别评测数据集上以85.22%的宏平均F1值超越了此前的最佳成绩,验证了融合语义信息和句法信息对于隐喻识别任务具有重要作用。
Abstract
Metaphor is a special phenomenon in human languages. As for Metaphor Detection in Chinese, we propose a SaGE (Syntax-aware GCN with ELECTRA) method inspired by linguistics. SaGE utilizes ELECTRA and Transformer encoder to extract the semantic feature of a sentence, and the GCN to extract syntactic feature through a graph constructed by dependency parsing result. The model concatenates the two features to detect metaphors. SaGE obatins 85.22% macro-F1 score, a substantial improvement over the best reported score in CCL 2018 Chinese Metaphor Detection Task Dataset.
关键词
隐喻识别 /
ELECTRA /
图卷积神经网络 /
依存句法
{{custom_keyword}} /
Key words
metaphor detection /
ELECTRA /
GCN /
dependency parsing
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] LAKOFF G, JOHNSON M. Metaphors we live by[M]. Chicago: University of Chicago Press, 1980.
[2] 贾玉祥, 俞士汶. 基于词典的名词性隐喻识别[J]. 中文信息学报, 2011, 25(02):99-104.
[3] 曾华琳, 周昌乐, 陈毅东,等. 基于特征自动选择方法的汉语隐喻计算[J]. 厦门大学学报(自然科学版), 2016, 55(03):406-412.
[4] 张冬瑜, 崔紫娟, 李映, 等. 基于Transformer和BERT的名词隐喻识别[J]. 数据分析与知识发现, 2020, 4(04):100-108.
[5] RUI MAO, LIN CH-H, GUERIN F. End-to-end sequential metaphor identification inspired by linguistic theories[C]//Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, 2019:3888-3898.
[6] MU J, YANNAKOUDAKIS H, SHUTOVA E. Learning outside the box: Discourse-level features improve metaphor identication[C]//Proceedings of NAACL HLT, 2019: 596-601.
[7] SU CH D, FUKUMOTO F, HUANG X X,et al. DeepMet: A reading comprehension paradigm for token-level metaphor detection[C]//Proceedings of the 2nd Workshop on Figurative Language Processing, 2020:30-39.
[8] GONG H Y, GUPTA K, JAIN A, et al. IlliniMet: Illinois system for metaphor detection with contextual and linguistic information[C]//Proceedings of the 2nd Workshop on Figurative Language Processing, 2020:146-153.
[9] TAI K S, SOCHER R, MANNING C D. Improved semantic representations from tree-structured long short-term memory networks[C]//Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing, 2015:1556-1566.
[10] MARCHEGGIANI D, TITOV I. Encoding sentences with graph convolutional networks for semantic role labeling[C]//Proceedings of the Conference on Empirical Methods in Natural Language Processing, 2017: 1506-1515.
[11] JI T, WU Y B, LAN M. Graph-based dependency parsing with graph neural networks[C]//Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, 2019:2475-2485.
[12] CLARK K, MINH-THANG L, QUOC V L,et al. ELECTRA: Pretraining text encoders as discriminators rather than generators[J]. arXiv preprint. 2020, arXiv:2003.10555.
[13] VASWANI A, SHAZEER N, PARMAR N,et al. Attention is all you need[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems, 2017:6000-6010.
[14] GILMER J, SCHOENHOLZ S S, RILEY P F, et al. Neural message passing for quantum chemistry[C]//Proceedings of the 34th International Conference on Machine Learning, 2017: 1263-1272.
[15] 束定芳. 论隐喻的运作机制[J]. 外语教学与研究, 2002(02):98-106,160.
[16] 束定芳. 论隐喻的基本类型及句法和语义特征[J]. 外国语(上海外国语大学学报), 2000(01):20-28.
[17] JACOB D, CHANG M W, LEE K,et al. BERT: Pre-training of deep bidirectional transformers for language understanding[J]. arXiv preprint. 2018, arXiv:1810.04805.
[18] SUN Y, WANG SH H, LI Y K,et al. Ernie: Enhanced representation through knowledge integration[J].arXiv preprint. 2019, arXiv: 1907.12412.
[19] YANG ZH L, DAI Z H, YANG Y M, et al. XLNet: Generalized autoregressive pretraining for language understanding[J]. arXiv preprint. 2019, arXiv:1906.08237.
[20] CUI Y M, CHE W X, LIU T, et al. Revisiting pre-trained models for Chinese natural language processing[C]//Proceedings of the Conference on Empirical Methods in Natural Language Processing: Findings, 2020: 657-668.
[21] 赵晓妮. 基于比喻性修辞的文本情感分析研究[D]. 哈尔滨:哈尔滨工业大学硕士学位论文, 2019.
[22] CHE W X, FENG Y L, QIN L B, et al. N-LTP: A open-source neural Chinese language technology platform with pretrained models[J]. arXiv preprint. 2020, arXiv:2009.11616.
[23] WANG M J, ZHENG D, YE Z H, et al. Deep graph library: A graph-centric, highly-performant package for graph neural networks[J]. arXiv preprint. 2020, arXiv:1909.01315.
[24] SULLIVAN K. Frame and constructions in metaphoric language[M]. Amsterdan: John Benjamins Publishing Company, 2013.
[25] HWANG J. Identification and representation of caused-motion constructions[D]. University of Colorado, 2014.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
基金
清华大学人文社科振兴项目(2019THZWJC38);教育部人文社会科学研究一般项目(17YJAZH056);国家社会科学基金(18ZDA238);百度网讯科技有限公司项目“基于深度学习框架PaddlePaddle的开源教程和案例建设”
{{custom_fund}}
PDF(1780 KB)
