基于图卷积神经网络的古汉语分词研究

doi:10.3772/j.issn.1000-0135.2023.06.009

情报学报

2023, Vol. 42

Issue (6): 740-750 DOI: 10.3772/j.issn.1000-0135.2023.06.009

情报技术与应用

本期目录 | 过刊浏览 | 高级检索

基于图卷积神经网络的古汉语分词研究

唐雪梅^1,2, 苏祺^3,2,4, 王军^1,2,4, 杨浩^2,4

1.北京大学信息管理系,北京 100871
2.北京大学数字人文研究中心,北京 100871
3.北京大学外国语学院,北京 100871
4.北京大学人工智能研究院,北京 100871

Ancient Chinese Word Segmentation Based on Graph Convolutional Neural Network

Tang Xuemei^1,2, Su Qi^3,2,4, Wang Jun^1,2,4, Yang Hao^2,4

1.Department of Information Management, Peking University, Beijing 100871
2.Digital Humanities Center of Peking University, Beijing 100871
3.School of Foreign Languages, Peking University, Beijing 100871
4.Institute for Artificial Intelligence, Peking University, Beijing 100871

摘要
图/表
参考文献
相关文章 (1)

全文: PDF (1230 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要古汉语的语法有省略、语序倒置的特点，词法有词类活用、代词名词丰富的特点，这些特点增加了古汉语分词的难度，并带来严重的out-of-vocabulary（OOV）问题。目前，深度学习方法已被广泛地应用在古汉语分词任务中并取得了成功，但是这些研究更关注的是如何提高分词效果，忽视了分词任务中的一大挑战，即OOV问题。因此，本文提出了一种基于图卷积神经网络的古汉语分词框架，通过结合预训练语言模型和图卷积神经网络，将外部知识融合到神经网络模型中来提高分词性能并缓解OOV问题。在《左传》《战国策》和《儒林外史》3个古汉语分词数据集上的研究结果显示，本文模型提高了3个数据集的分词表现。进一步的研究分析证明，本文模型能够有效地融合词典和N-gram信息；特别是N-gram有助于缓解OOV问题。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	唐雪梅
	苏祺
	王军
	杨浩

关键词 ：古汉语, 汉语分词, 图卷积神经网络, 预训练语言模型, BERT（bidirectional encoder representations from transformers）

收稿日期: 2022-07-14

基金资助:国家自然科学基金国际重点合作项目“中国儒家学术史知识图谱构建研究”（72010107003）。

作者简介: 唐雪梅，女，1995年生，博士研究生，主要研究领域为数字人文和自然语言处理；苏祺，女，1979年生，博士，研究员，长聘副教授，博士生导师，主要研究领域为自然语言处理、计算语言学、语料库语言学、数字人文，E-mail：sukia@pku.edu.cn；王军，男，1968年生，博士，教授，博士生导师，主要研究领域为知识图谱、数据分析、用户行为研究、数字人文；杨浩，男，1981年生，博士，副研究员，主要研究领域为中国哲学、数字人文；

引用本文:

唐雪梅, 苏祺, 王军, 杨浩. 基于图卷积神经网络的古汉语分词研究[J]. 情报学报, 2023, 42(6): 740-750.
Tang Xuemei, Su Qi, Wang Jun, Yang Hao. Ancient Chinese Word Segmentation Based on Graph Convolutional Neural Network. 情报学报, 2023, 42(6): 740-750.

链接本文:

https://qbxb.istic.ac.cn/CN/10.3772/j.issn.1000-0135.2023.06.009 或 https://qbxb.istic.ac.cn/CN/Y2023/V42/I6/740

1 国家技术监督局. 中华人民共和国国家标准: 信息处理用现代汉语分词规范(GB/T 13715—92)[S]. 北京: 中国标准出版社, 1993.
2 张琪, 江川, 纪有书, 等. 面向多领域先秦典籍的分词词性一体化自动标注模型构建[J]. 数据分析与知识发现, 2021, 5(3): 2-11.
3 高毅. 基于BERT预训练模型的古汉语自动分词方法研究[J]. 电子设计工程, 2021, 29(22): 28-32.
4 刘畅, 王东波, 胡昊天, 等. 面向数字人文的融合外部特征的典籍自动分词研究——以SikuBERT预训练模型为例[J]. 图书馆论坛, 2022, 42(6): 44-54.
5 Yao L, Mao C S, Luo Y. Graph convolutional networks for text classification[C]// Proceedings of the Thirty-Third AAAI Conference on Artificial Intelligence and Thirty-First Innovative Applications of Artificial Intelligence Conference and Ninth AAAI Symposium on Educational Advances in Artificial Intelligence. Palo Alto: AAAI Press, 2019: 7370-7377.
6 Zhao L, Zhang A L, Liu Y, et al. Encoding multi-granularity structural information for joint Chinese word segmentation and POS tagging[J]. Pattern Recognition Letters, 2020, 138: 163-169.
7 Hu L M, Yang T C, Shi C, et al. Heterogeneous graph attention networks for semi-supervised short text classification[C]// Proceedings of the 9th International Joint Conference on Natural Language Processing. Stroudsburg: Association for Computational Linguistics, 2019: 4820-4829.
8 Bastings J, Titov I, Aziz W, et al. Graph convolutional encoders for syntax-aware neural machine translation[C]// Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing. Stroudsburg: Association for Computational Linguistics, 2017: 1957-1967.
9 Liu J X, Wu F Z, Wu C H, et al. Neural Chinese word segmentation with dictionary knowledge[C]// Proceedings of the CCF International Conference on Natural Language Processing and Chinese Computing. Cham: Springer, 2018: 80-91.
10 Ma J, Ganchev K, Weiss D. State-of-the-art Chinese word segmentation with Bi-LSTMs[C]// Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing. Stroudsburg: Association for Computational Linguistics, 2018: 4902-4908.
11 Zhang M S, Zhang Y, Fu G H. Transition-based neural word segmentation[C]// Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics. Stroudsburg: Association for Computational Linguistics, 2016: 421-431.
12 Chen X C, Qiu X P, Zhu C X, et al. Long short-term memory neural networks for Chinese word segmentation[C]// Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing. Stroudsburg: Association for Computational Linguistics, 2015: 1197-1206.
13 Qiu X P, Pei H Z, Yan H, et al. A concise model for multi-criteria Chinese word segmentation with transformer encoder[C]// Proceedings of the Conference on the Findings of the Association for Computational Linguistics: EMNLP 2020. Stroudsburg: Association for Computational Linguistics, 2020: 2887-2897.
14 Margatina K, Baziotis C, Potamianos A. Attention-based conditioning methods for external knowledge integration[C]// Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics. Stroudsburg: Association for Computational Linguistics, 2019: 3944-3951.
15 Ding N, Long D, Xu G, et al. Coupling distant annotation and adversarial training for cross-domain Chinese word segmentation[C]// Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics. Stroudsburg: Association for Computational Linguistics, 2020: 6662-6671.
16 Liu W, Fu X Y, Zhang Y, et al. Lexicon enhanced Chinese sequence labeling using BERT adapter[C]// Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing. Stroudsburg: Association for Computational Linguistics, 2021: 5847-5858.
17 Tian Y, Song Y, Xia F, et al. Improving Chinese word segmentation with wordhood memory networks[C]// Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics. Stroudsburg: Association for Computational Linguistics, 2020: 8274-8285.
18 Tian Y, Song Y, Xia F. Joint Chinese word segmentation and part-of-speech tagging via multi-channel attention of character N-grams[C]// Proceedings of the 28th International Conference on Computational Linguistics. Barcelona: International Committee on Computational Linguistics, 2020: 2073-2084.
19 郭辉, 苏中义, 王文, 等. 一种改进的MM分词算法[J]. 微型电脑应用, 2002(1): 13-15, 2.
20 邱冰, 皇甫娟. 基于中文信息处理的古代汉语分词研究[J]. 微计算机信息, 2008, 24(24): 100-102.
21 王嘉灵. 以《汉书》为例的中古汉语自动分词[D]. 南京: 南京师范大学, 2014.
22 梁社会, 陈小荷. 先秦文献《孟子》自动分词方法研究[J]. 南京师范大学文学院学报, 2013(3): 175-182.
23 高嘉琦, 赵庆聪. 基于新词发现的古典文学作品分词方法研究[J]. 计算机技术与发展, 2021, 31(9): 178-181, 207.
24 邢付贵, 朱廷劭. 基于大规模语料库的古文词典构建及分词技术研究[J]. 中文信息学报, 2021, 35(7): 41-46.
25 钱智勇, 周建忠, 童国平, 等. 基于HMM的楚辞自动分词标注研究[J]. 图书情报工作, 2014, 58(4): 105-110.
26 王晓玉, 李斌. 基于CRFs和词典信息的中古汉语自动分词[J]. 数据分析与知识发现, 2017, 1(5): 62-70.
27 杨世超. 古汉语分词与词性标注方法研究[D]. 唐山: 华北理工大学, 2018.
28 程宁, 李斌, 葛四嘉, 等. 基于BiLSTM-CRF的古汉语自动断句与词法分析一体化研究[J]. 中文信息学报, 2020, 34(4): 1-9.
29 俞敬松, 魏一, 张永伟. 基于BERT的古文断句研究与应用[J]. 中文信息学报, 2019, 33(11): 57-63.
30 Zhou Z H, Li M. Tri-training: exploiting unlabeled data using three classifiers[J]. IEEE Transactions on Knowledge and Data Engineering, 2005, 17(11): 1529-1541.
31 Nguyen M V, Min B, Dernoncourt F, et al. Joint extraction of entities, relations, and events via modeling inter-instance and inter-label dependencies[C]// Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. Stroudsburg: Association for Computational Linguistics, 2022: 4363-4374.
32 Defferrard M, Bresson X, Vandergheynst P. Convolutional neural networks on graphs with fast localized spectral filtering[C]// Proceedings of the 30th Annual Conference on Neural Information Processing Systems. Cambridge: MIT Press, 2016: 3844-3852.
33 Marcheggiani D, Titov I. Encoding sentences with graph convolutional networks for semantic role labeling[C]// Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing. Stroudsburg: Association for Computational Linguistics, 2017: 1507-1516.
34 Du J, Mi W, Du X. Chinese word segmentation in electronic medical record text via graph neural network-bidirectional LSTM-CRF model[C]// Proceedings of the 2020 IEEE International Conference on Bioinformatics and Biomedicine. Piscataway: IEEE, 2020: 985-989.
35 Huang K, Yu H, Liu J P, et al. Lexicon-based graph convolutional network for Chinese word segmentation[C]// Findings of the Association for Computational Linguistics: EMNLP 2021. Stroudsburg Association for Computational Linguistics, 2021: 2908-2917.
36 Devlin J, Chang M W, Lee K, et al. BERT: pre-training of deep bidirectional transformers for language understanding[C]// Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. Stroudsburg Association for Computational Linguistics, 2019: 4171-4186.
37 唐雪梅, 苏祺, 王军, 等. 基于预训练语言模型的繁体古文自动句读研究[C]// 第二十届中国计算语言学大会. 北京: 中国中文信息学会, 2021: 678-688.
38 Lafferty J, McCallum A, Pereira F. Conditional random fields: probabilistic models for segmenting and labeling sequence data[C]// Proceedings of the Eighteenth International Conference on Machine Learning. San Francisco: Morgan Kaufmann Publishers, 2001: 282-289.
39 Feng H D, Chen K, Deng X T, et al. Accessor variety criteria for Chinese word extraction[J]. Computational Linguistics, 2004, 30(1): 75-93.
40 王东波, 刘畅, 朱子赫, 等. SikuBERT与SikuRoBERTa: 面向数字人文的《四库全书》预训练模型构建及应用研究[J]. 图书馆论坛, 2022, 42(6): 31-43.
41 Wei X Y, Liu W H, Qing Z, et al. Glyph features matter: a multimodal solution for EvaHan in LT4HALA2022[C]// Proceedings of the Second Workshop on Language Technologies for Historical and Ancient Languages. Marseille: European Language Resources Association, 2022: 178-182. 责任编辑王克平）