基于汉字多模信息与象形视觉对齐增强的古籍文本命名实体识别研究

doi:10.3772/j.issn.1000-0135.2025.04.006

情报学报

2025, Vol. 44

Issue (4): 452-465 DOI: 10.3772/j.issn.1000-0135.2025.04.006

情报技术与应用

本期目录 | 过刊浏览 | 高级检索

基于汉字多模信息与象形视觉对齐增强的古籍文本命名实体识别研究

郑旭辉^1,2, 王昊^1,2, 裘靖文^1,2

1.南京大学信息管理学院，南京 210023
2.江苏省数据工程与知识服务重点实验室，南京 210023

Named Entity Recognition of Ancient Texts Based on the Enhancement of Multimodal Information from Chinese Characters and Pictographic Visual Alignment

Zheng Xuhui^1,2, Wang Hao^1,2, Qiu Jingwen^1,2

1.School of Information Management, Nanjing University, Nanjing 210023
2.Jiangsu Key Laboratory of Data Engineering and Knowledge Service, Nanjing 210023

摘要
图/表
参考文献
相关文章 (4)

全文: PDF (4266 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要古籍的语义解析与人文计算是建设文化强国的重要组成部分，而古籍文本命名实体识别（named entity recognition，NER）是开展后续古籍知识发现与组织的前提和基础，设计一种适用于简体化文言文特性的命名实体识别模型具有重要的研究意义。汉字本身具有大量象形特征的视觉信息与发音信息，这些更贴合汉字发展历史的知识能够为识别古籍中的实体提供更多的信息以提高模型性能。基于此，本文构建了基于多模态汉字象形表示的GMAE-NER（guwen multi-information alignment enhanced NER）模型，该模型创新性地提出了汉字象形层面里图像与笔画信息的多模态特征处理和对齐方法，实现了将BERT（bidirectional encoder representations from transformers）表征与汉字视觉信息、发音信息相融合，有效增强了古籍文本命名实体识别的效果。本文将模型在纪传体史书《后汉书》上进行了大量的实验与对比，发现相较于基线模型，GMAE-NER在各个类别实体识别的F1指标上均得到了1.32~15.00个百分点的提升，并且能更好地识别出古籍文本中重叠表述的实体，消融分析结果也充分证明了该模型中视觉编码、发音编码与特征融合模块的有效性。

	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
	郑旭辉
	王昊
	裘靖文

关键词 ：古籍文本, 中文命名实体识别, 汉字字形, 汉字发音, 跨模态交互融合

收稿日期: 2024-04-07

基金资助:国家自然科学基金面上项目“关联数据驱动下我国非遗文本的语义解析与人文计算研究”（72074108）。

作者简介: 郑旭辉，男，2000年生，硕士研究生，主要研究领域为自然语言处理；王昊，通信作者，男，1981年生，博士，教授，博士生导师，主要研究领域为数字人文与知识组织，E-mail：ywhaowang@nju.edu.cn；裘靖文，女，1999年生，博士研究生，主要研究领域为数据挖掘与信息分析；

引用本文:

郑旭辉, 王昊, 裘靖文. 基于汉字多模信息与象形视觉对齐增强的古籍文本命名实体识别研究[J]. 情报学报, 2025, 44(4): 452-465.
Zheng Xuhui, Wang Hao, Qiu Jingwen. Named Entity Recognition of Ancient Texts Based on the Enhancement of Multimodal Information from Chinese Characters and Pictographic Visual Alignment. 情报学报, 2025, 44(4): 452-465.

链接本文:

https://qbxb.istic.ac.cn/CN/10.3772/j.issn.1000-0135.2025.04.006 或 https://qbxb.istic.ac.cn/CN/Y2025/V44/I4/452

1 推进新时代古籍工作[N]. 人民日报, 2022-04-12(1).
2 欧阳剑. 数字人文应用服务中的数据版权风险及防范策略[J]. 中国图书馆学报, 2023, 49(1): 118-128.
3 黄水清, 王晓光, 夏翠娟, 等. 推进新时代古籍工作, 加快创新智能化发展[J]. 农业图书情报学报, 2022, 34(5): 4-20.
4 刘浏, 王东波. 命名实体识别研究综述[J]. 情报学报, 2018, 37(3): 329-340.
5 Yadav V, Sharp R, Bethard S. Deep affix features improve neural named entity recognizers[C]// Proceedings of the Seventh Joint Conference on Lexical and Computational Semantics. Stroudsburg: Association for Computational Linguistics, 2018: 167-172.
6 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.
7 Radford A, Narasimhan K, Salimans T, et al. Improving language understanding by generative pre-training[OL]. https://cdn.openai.com/research-covers/language-unsupervised/language_understanding_paper.pdf.
8 Norman J. Chinese[M]. Cambridge: Cambridge University Press, 1988.
9 杜悦, 王东波, 江川, 等. 数字人文下的典籍深度学习实体自动识别模型构建及应用研究[J]. 图书情报工作, 2021, 65(3): 100-108.
10 屈倩倩, 阚红星. 基于BERT-BiLSTM-CRF的中医文本命名实体识别[J]. 电子设计工程, 2021, 29(19): 40-43, 48.
11 林立涛, 王东波, 刘江峰, 等. 数字人文视域下典籍动物命名实体识别研究——以SikuBERT预训练模型为例[J]. 图书馆论坛, 2022, 42(10): 42-50.
12 刘江峰, 冯钰童, 王东波, 等. 数字人文视域下SikuBERT增强的史籍实体识别研究[J]. 图书馆论坛, 2022, 42(10): 61-72.
13 范涛, 王昊, 陈玥彤. 基于深度迁移学习的地方志多模态命名实体识别研究[J]. 情报学报, 2022, 41(4): 412-423.
14 谢靖, 刘江峰, 王东波. 古代中国医学文献的命名实体识别研究——以Flat-Lattice增强的SikuBERT预训练模型为例[J]. 图书馆论坛, 2022, 42(10): 51-60.
15 武帅, 杨秀璋, 何琳, 等. 基于句法特征和BERT-BiLSTM-MHA-CRF的细粒度古籍实体识别研究[J]. 数据分析与知识发现2024, 8(12): 136-148.
16 Sun Y M, Lin L, Yang N, et al. Radical-enhanced Chinese character embedding[C]// Proceedings of the International Conference on Neural Information Processing. Cham: Springer, 2014: 279-286.
17 Shi X L, Zhai J J, Yang X D, et al. Radical embedding: delving deeper to Chinese radicals[C]// Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing. Stroudsburg: Association for Computational Linguistics, 2015: 594-598.
18 Liu F, Lu H, Lo C, et al. Learning character-level compositionality with visual features[C]// Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics. Stroudsburg: Association for Computational Linguistics, 2017: 2059-2068.
19 Dai F Z, Cai Z. Glyph-aware embedding of Chinese characters[C]// Proceedings of the First Workshop on Subword and Character Level Models in NLP. Stroudsburg: Association for Computational Linguistics, 2017: 64-69.
20 Su T R, Lee H Y. Learning Chinese word representations from glyphs of characters[C]// Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing. Stroudsburg: Association for Computational Linguistics, 2017: 264-273.
21 Meng Y X, Wu W, Wang F, et al. Glyce: glyph-vectors for Chinese character representations[C]// Proceedings of the 32nd Conference on Neural Information Processing Systems. Red Hook: Curran Associates, 2020: 2723-2734.
22 Sun Z J, Li X Y, Sun X F, et al. ChineseBERT: Chinese pretraining enhanced by glyph and pinyin information[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: 2065-2075.
23 Xuan Z Y, Bao R, Jiang S Y. FGN: fusion glyph network for Chinese named entity recognition[C]// Proceedings of the Conference on Knowledge Graph and Semantic Computing: Knowledge Graph and Cognitive Intelligence. Singapore: Spring, 2021: 28-40.
24 Gu R M, Wang T, Deng J F, et al. Improving Chinese named entity recognition by interactive fusion of contextual representation and glyph representation[J]. Applied Sciences, 2023, 13(7): 4299.
25 Guo X C, Lu S H, Tang Z, et al. CG-ANER: enhanced contextual embeddings and glyph features-based agricultural named entity recognition[J]. Computers and Electronics in Agriculture, 2022, 194: 106776.
26 Li J T, Meng K. MFE-NER: multi-feature fusion embedding for Chinese named entity recognition[C]// Proceedings of the 23rd China National Conference on Chinese Computational Linguistics. Singapore: Springer, 2025: 191-204.
27 Lv C, Zhang H, Du X K, et al. StyleBERT: Chinese pretraining by font style information[C]// Proceedings of the 10th IEEE Joint International Information Technology and Artificial Intelligence Conference. Piscataway: IEEE, 2022: 646-652.
28 Wu S, Song X N, Feng Z H. MECT: multi-metadata embedding based cross-transformer for Chinese named entity recognition[C]// Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing. Online. Stroudsburg: Association for Computational Linguistics, 2021: 1529-1539.
29 Zhang B H, Cai J H, Zhang H P, et al. VisPhone: Chinese named entity recognition model enhanced by visual and phonetic features[J]. Information Processing & Management, 2023, 60(3): 103314.
30 Wang Y H, Chen H T, Tang Y H, et al. PanGu-π: enhancing language model architectures via nonlinearity compensation[OL]. (2023-12-27). https://arxiv.org/pdf/2312.17276.
31 Li J Y, Fei H, Liu J, et al. Unified named entity recognition as word-word relation classification[J]. Proceedings of the AAAI Conference on Artificial Intelligence, 2022, 36(10): 10965-10973. 责任编辑魏瑞斌）