[Purpose/Significance] The reference resolution is one of the key steps to acquire knowledge and construct the knowledge map of ancient books. However, there is a lack of published corpora of referential relations of ancient TCM literature, which makes it difficult to carry out the research of reference resolution. Based on this, this study takes ancient literature on stroke and abdominal pain during pregnancy as examples, constructs a corpus of referential relations, and conducts experiments on reference resolution methods to provide references for the following research. [Method/Process] This paper selected ancient literature records of stroke and pregnancy abdominal pain from the “Chinese Medical Code”, defined and annotated three types of pronouns, two types of antecedents and one type of referential relations, and constructed a corpus through manual annotation and evaluated its quality. In the experimental stage, the reference resolution task was divided into two subtasks: referent recognition and referential relations prediction. The experimental group selected the reference resolution model BERT-BiLSTM-CRF based on the pre-trained model BERT, and compared it with the BiLSTM-CRF and CRF models. [Result/Conclusion] The average score of corpus inter-annotator agreement (IAA) is 0.87, suggesting that the corpus quality is good. The BERT-BiLSTM-CRF model is significantly better than the other two models in the reference resolution task. In general, it is feasible to use deep learning technology to perform the reference resolution of ancient TCM literature, and the pre-trained model BERT can improve the model performance. Future research will need to further increase the scale of the corpus and train large language models to help the research of knowledge discovery on ancient TCM literature.
[1] 程小恩, 温川飙, 许强, 等. 基于中医药人工智能技术探讨中医药大数据的典型特征[J]. 世界科学技术—中医药现代化, 2020, 22(4): 1243-1248. (CHENG X, WEN C B, XU Q et al. Development demand of artificial intelligence based on traditional Chinese medicine to study on the typical of the big data of traditional Chinese medicine[J]. Modernization of traditional Chinese medieine and materia materia-world seience and technology, 2020, 22(4): 1243-1248.)
[2] ZHANG T, HUANG Z, WANG Y, et al. Information extraction from the text data on traditional Chinese medicine: a review on tasks, challenges, and methods from 2010 to 2021[J]. Evidence-based complementary and alternative medicine, 2022, 2022(14): 1-19.
[3] 孙超, 张文博. 中医古籍文本术语命名实体识别的研究进展与挑战[J]. 中华中医药杂志, 2021, 36(11): 6843-6845. (SUN C, ZHANG B W. Research progress and challenges of named entity recognition of terms inancient Chinese medicine books[J]. China journal of traditional Chinese medicine and pharmacy, 2021, 36(11): 6843-6845.)
[4] 刘丽. 中医经方知识图谱的研究与构建[D]. 济南: 山东中医药大学, 2022. (LIU L. Study and construction of classical formula knowledge graph[D]. Jinan: Shandong University of Traditional Chinese Medicine, 2022.)
[5] 李娜. 面向方志类古籍的多类型命名实体联合自动识别模型构建[J]. 图书馆论坛, 2021, 41(12): 113-123. (LI N. Construction of automatic recognition model of multi-type named entities for local gazetteers[J]. Library tribune, 2021, 41(12): 113-123.)
[6] 曾子玲. 基于古籍的阴虚证知识图谱构建及其关键技术研究[D]. 北京: 中国中医科学院, 2022. (ZENG Z. Construction of knowledge map of yin deficiency syndrome based on ancient books and research on key technologies[D]. Beijing: China Academy of Chinese Medical Sciences, 2022.)
[7] 包振山, 宋秉彦, 张文博, 等. 基于半监督学习和规则相结合的中医古籍命名实体识别研究[J]. 中文信息学报, 2022, 36(6): 90-100. (BAO Z S, SONG B Y, ZHANG W B, et al. Named entity recognition in traditional Chinese mdicine books combining semi-supervised learning and rule-based approach[J]. Journal of Chinese information processing, 2022, 36(6): 90-100.)
[8] XI W, GU P. The language of ancient Chinese medicine[J]. The Chinese journal for the history of science and technology, 2020, 41(3): 298-305.
[9] 纪晓丽, 张辉, 李爱军. 神经语言学视角下语言学热点问题研究——以《科学》杂志《语言与大脑》专刊为例[J]. 西安外国语大学学报, 2021, 29(2): 1-7. (JI X L, ZHANG H, LI A J. Research on hot linguistic issues from the perspective of neurolinguistics: taking the special issue “language and brain” of science magazine as an example[J]. Journal of Xi’an International Studies University, 2021, 29(2): 1-7.)
[10] MASSIMO P, ROLAND S, YANNICK V. Anaphora resolution[M]. 1st ed. Berlin: Springer, 2016.
[11] RHEA S, SOUJANYA P, ERIK C, et al. Anaphora and coreference resolution: a review[J]. Information fusion, 2020, 59(2020): 139-162.
[12] CONVERSE S P. Developing guidelines for the annotation of anaphors in the Chinese Treebank[C]//Proceedings of the first SIGHAN workshop on Chinese language processing. Philadelphia: University of Pennsylvania, 2002: 1-7.
[13] DODDINGTON G, MITCHELL A, PRZYBOCKI M, et al. The automatic content extraction (ace) program-tasks, data, and evaluation[C]// Proceedings of LREC. Lisbon: Researchr, 2004: 837-840.
[14] PRADHAN S, MOSCHITTI A, XUE N, et al. CoNLL-2012 shared task: Modeling multilingual unrestricted coreference in OntoNotes[C]//Proceedings of the Joint Conference on EMNLP and CoNLL-Shared Task. Jeju Island: Association for Computational Linguistics, 2012: 1-40.
[15] PENG Y, ZHANG Y, HUANG S, et al. Resolution of personal pronoun anaphora in Chinese micro-blog[C]//Lecture Notes in Computer Science. Cham: Springer, 2018: 592-605.
[16] U.S. National Institute of Standards and Technology (NIST). TAC knowledge base population (KBP) 2017[EB/OL]. [2025-05-21]. https://tac.nist.gov/2017/KBP/.
[17] Chinese Language Understanding Evaluation Benchmark: datasets, baselines, pre-trained models, corpus and leaderboard [EB/OL]. [2025-05-21]. https://www.cluebenchmarks.com/.
[18] LI S N, QU W G, WEI T X, et al. A survey of chinese anaphora resolution[C]//Lecture Notes in Computer Science. 7th International Conference. Dublin: Springer, 2021:180-192.
[19] MASSIMO P, YU J T, SILVIU P, et al. Computational models of anaphora[J]. Annual review of linguistics, 2023, 9: 561-587.
[20] 饶志强. 基于领域规则和深度学习的文本信息提取[D]. 西安: 西安电子科技大学, 2017. (RAO Z Q. Text information extraction based on domain rules and deep learning[D]. Xi’an: Xidian University, 2017.)
[21] JIANG F, COHN T. Incorporating syntax and semantics in coreference resolution with heterogeneous graph attention network[C]//Proceedings of the 2021 conference of the North American Chapter of the Association for Computational Linguistics: Human language technologies. Mexico City: North American Chapter of the Association for Computational Linguistics, 2021: 1584-1591.
[22] 陈诗, 王东波, 黄水清. 数字人文下的典籍人称代词指代消解研究[J]. 情报理论与实践, 2021, 44(10): 165-172. (CHEN S, WANG D B, HUANG S Q. Research on the resolution of personal pronoun in classical books under the digital humanism[J]. Information studies: theory & application, 2021, 44(10): 165-172.)
[23] JOSHI M, LEVY O, WELD D S, et al. BERT for coreference resolution: baselines and analysis[EB/OL]. [2025-05-21]. https://arxiv.org/abs/1908.09091.
[24] 陈雨, 玄宇航, 张玉志. 基于深度学习和指代消解的中文人名识别[J]. 数据与计算发展前沿, 2022, 4(2): 63-73. (CHEN Y, XUAN Y H, ZHANG Y Z. Research on Chinese name recognition based on deep learning and coreference resolution[J]. Frontiers of data & computing, 2022, 4(2): 63-73.)
[25] ALEXANDER S, TATIANA L. Knowledge fusion patterns: a survey[J]. Information fusion, 2019, 52(2019): 31-40.
[26] ZHAO X J, JIA Y, LI A, et al. Multi-source knowledge fusion: a survey[J]. World Wide Web, 2020, 23(4): 2567-2592.
[27] HE B, DONG B, GUAN Y er al. Building a comprehensive syntactic and semantic corpus of Chinese clinical texts[J]. Journal of biomedical informatics, 2017, 69: 203-217.
[28] ZHOU X, PENG Y, LIU B. Text mining for traditional Chinese medical knowledge discovery: a survey[J]. Journal of biomedical informatics, 2010, 43(4): 650-660.
[29] DING P S, PACKARD J. New Approaches to Chinese Word Formation: Morphology, Phonology, and the Lexicon in Modern and Ancient Chinese[J]. Language, 1997, 75(2): 396.
[30] KUSUM L, PARDEEP S, KANLESH D. A comprehensive review on feature set used for anaphora resolution[J]. Artificial intelligence review, 2021, 54(4): 2917-3006.
[31] 裘沛然. 中华医典[CD]. 长沙: 湖南电子音像出版社, 2014. (QIU P R. Chinese medical canon[CD]. Changsha: Hunan Electronic & Audio-Video Press, 2014.)
[32] NADKARNI P, OHNO-MACHADO L, CHAPMAN W. Natural language processing: an introduction[J]. Journal of the American Medical Informatics Association, 2011, 18(5): 544-551.
[33] ZHANG T, WANG Y, WANG X, et al. Constructing fine-grained entity recognition corpora based on clinical records of traditional Chinese medicine[J]. BMC medical informatics and decision making, 2020, 20(1): 1-17.
[34] COHEN J. A coefficient of agreement for nominal scales[J]. Educational and psychological measurement, 1960, 20(1): 37-46.
[35] TANG W, HU J, ZHANG H et al. Kappa coefficient: a popular measure of rater agreement[J]. Shanghai arch psychiatry, 2015, 27(1): 62-67.
[36] CARLETTA J. Assessing agreement on classification tasks: the kappa statistic[J]. Computational linguistics, 1996, 22(2): 249-254.
