命名实体识别方法研究综述

doi:10.3778/j.issn.1673-9418.2112109

计算机科学与探索 ›› 2022, Vol. 16 ›› Issue (9): 1954-1968.DOI: 10.3778/j.issn.1673-9418.2112109

命名实体识别方法研究综述

李冬梅¹^,², 罗斯斯¹^,², 张小平³^,⁺(), 许福¹^,²

1.北京林业大学信息学院,北京 100083
2.国家林业和草原局林业智能信息处理工程技术研究中心,北京 100083
3.中国中医科学院中医药信息研究所,北京 100700

收稿日期:2021-12-29 修回日期:2022-04-29 出版日期:2022-09-01 发布日期:2022-09-15
通讯作者: + E-mail: xiao_ping_zhang@139.com
作者简介:李冬梅（1972—）,女,博士,教授,主要研究方向为自然语言处理、知识图谱。
罗斯斯（1993—）,女,硕士研究生,主要研究方向为自然语言处理、知识图谱。
张小平（1969—）,女,博士,正高级工程师,主要研究方向为数据挖掘、人工智能。
许福（1979—）,男,博士,教授,主要研究方向为遥感信息处理、智慧林业、智慧园林。
基金资助:
中央级公益性科研院所基本科研业务费专项资金(ZZ140319-W);国家自然科学基金(61772078)

Review on Named Entity Recognition

LI Dongmei¹^,², LUO Sisi¹^,², ZHANG Xiaoping³^,⁺(), XU Fu¹^,²

1. School of Information Science and Technology, Beijing Forestry University, Beijing 100083, China
2. Engineering Research Center for Forestry-Oriented Intelligent Information Processing, National Forestry and Grass-land Administration, Beijing 100083, China
3. Institute of Information on Traditional Chinese Medicine, China Academy of Chinese Medical Sciences, Beijing 100700, China

Received:2021-12-29 Revised:2022-04-29 Online:2022-09-01 Published:2022-09-15
About author:LI Dongmei, born in 1972, Ph.D., professor. Her research interests include natural language processing and knowledge graph.
LUO Sisi, born in 1993, M.S. candidate. Her re-search interests include natural language pro-cessing and knowledge graph.
ZHANG Xiaoping, born in 1969, Ph.D., profes-sorate senior engineer. Her research interests include data mining and artificial intelligence.
XU Fu, born in 1979, Ph.D., professor. His re-search interests include remote sensing informa-tion processing, smart forestry and smart garden.
Supported by:
Fundamental Research Funds for the Central Public Welfare Research Institutes(ZZ140319-W);National Natural Science Foundation of China(61772078)

摘要/Abstract

摘要：

在自然语言处理领域,命名实体识别是信息抽取的第一个关键环节。命名实体识别任务旨在从大量非结构化的文本中识别出命名实体并将其分类为预定义的类型,为关系抽取、文本摘要和机器翻译等自然语言处理任务提供基础支持。首先概述了命名实体识别的定义、研究难点和中文命名实体识别任务的特殊性,总结了命名实体识别任务中常用的中英文公共数据集和评估标准。然后根据命名实体识别的发展历程调研了现有的命名实体识别方法,主要为早期基于规则和词典的命名实体识别方法、基于统计机器学习的命名实体识别方法和基于深度学习的命名实体识别方法。归纳总结了每一种命名实体识别方法的关键思路、优缺点和具有代表性的模型,同时对各阶段的中文命名实体识别方法进行了总结。特别对最新的基于Transformer和基于提示学习的命名实体识别方法进行了综述,这两种细分类的方法是基于深度学习的命名实体识别方法中最先进的方法。最后总结了命名实体识别研究面临的挑战,并展望了未来的研究方向。

关键词: 自然语言处理, 命名实体识别, 机器学习, 深度学习, 关系抽取

Abstract:

In the field of natural language processing, named entity recognition is the first key step of information extraction. Named entity recognition task aims to recognize named entities from a large number of unstructured texts and classify them into predefined types. Named entity recognition provides basic support for many natural language processing tasks such as relationship extraction, text summarization, machine translation, etc. This paper first introduces the definition of named entity recognition, research difficulties, particularity of Chinese named entity recognition, and summarizes the common Chinese and English public datasets and evaluation criteria in named entity recognition tasks. Then, according to the development history of named entity recognition, the existing named entity recognition methods are investigated, which are the early named entity recognition methods based on rules and dictionaries, the named entity recognition methods based on statistic and machine learning, and the named entity recognition methods based on deep learning. This paper summarizes the key ideas, advantages and disadvan-tages and representative models of each named entity recognition method, and summarizes the Chinese named entity recognition methods in each stage. In particular, the latest named entity recognition based on Transformer and based on prompt learning are reviewed, which are state-of-the-art in deep learning-based named entity recognition methods. Finally, the challenges and future research trends of named entity recognition are discussed.

Key words: natural language processing, named entity recognition, machine learning, deep learning, relation extraction

中图分类号:

TP391

李冬梅, 罗斯斯, 张小平, 许福. 命名实体识别方法研究综述[J]. 计算机科学与探索, 2022, 16(9): 1954-1968.

LI Dongmei, LUO Sisi, ZHANG Xiaoping, XU Fu. Review on Named Entity Recognition[J]. Journal of Frontiers of Computer Science and Technology, 2022, 16(9): 1954-1968.

图/表 11

参考文献 73

[1]	GRISHMAN R, SUNDHEIM B. Message understanding con-ference-6: a brief history[C]// Proceedings of the 16th Inter-national Conference on Computational Linguistics, Copen-hagen, Aug 5-9, 1996. Stroudsburg: ACL, 1996: 466-471.
[2]	MERCHANT R, OKUROWSKI M E, CHINCHOR N. The multilingual entity task (MET) overview[C]// Proceedings of the Tipster Text Program Phase II, Vienna, May 6-8, 1996. Stroudsburg: ACL, 1996: 445-447.
[3]	SANG E F T K, DEMEULDER F. Introduction to the CoNLL-2003 shared task: language-independent named en-tity recognition[C]// Proceedings of the 7th Conference on Natural Language Learning, Held in Cooperation with HLT-NAACL 2003, Edmonton, May 31-Jun 1, 2003. Strouds-burg: ACL, 2003: 142-147.
[4]	DODDINGTON G R, MITCHELL A, PRZYBOCKI M A, et al. The automatic content extraction (ACE) program - tasks, data, and evaluation[C]// Proceedings of the 4th Inter-national Conference on Language Resources and Evalua-tion, Lisbon, May 26-28, 2004. Stroudsburg: ACL, 2004: 837-840.
[5]	孙镇, 王惠临. 命名实体识别研究进展综述[J]. 现代图书情报技术, 2010(6): 42-47.
	SUN Z, WANG H L. Overview on the advance of the resea-rch on named entity recognition[J]. New Technology of Library and Information Service, 2010(6): 42-47.
[6]	LI J, SUN A X, HAN J L, et al. A survey on deep learning for named entity recognition[J]. IEEE Transactions on Know-ledge and Data Engineering, 2022, 34(1): 50-70.
[7]	李猛, 李艳玲, 林民. 命名实体识别的迁移学习研究综述[J]. 计算机科学与探索, 2021, 15(2): 206-218. DOI
	LI M, LI Y L, LIN M. Review of transfer learning for named entity recognition[J]. Journal of Frontiers of Computer Science and Technology, 2021, 15(2): 206-218.
[8]	赵山, 罗睿, 蔡志平. 中文命名实体识别综述[J]. 计算机科学与探索, 2022, 16(2): 296-304. DOI
	ZHAO S, LUO R, CAI Z P. Survey of Chinese named en-tity recognition[J]. Journal of Frontiers of Computer Science and Technology, 2022, 16(2): 296-304. DOI
[9]	RAU L F. Extracting company names from text[C]// Procee-dings of the 7th IEEE Conference on Artificial Intelligence Application, Miami, Feb 24, 1991. Washington: IEEE Com-puter Society, 1991: 29-32.
[10]	KRUPKA G R. SRA: description of the SRA system as used for MUC-6[C]// Proceedings of the 6th Conference on Mess-age Understanding, Columbia, Nov 6-8, 1995. Stroudsburg: ACL, 1995: 221-235.
[11]	SHAALAN K, RAZA H. NERA: named entity recognition for Arabic[J]. Journal of the American Society for Informa-tion Science and Technology, 2009, 60(8): 1652-1663.
[12]	张小衡, 王玲玲. 中文机构名称的识别与分析[J]. 中文信息学报, 1997(4): 22-33.
	ZHANG X H, WANG L L. Identification and analysis of chinese organization and institution names[J]. Journal of Chinese Information Processing, 1997(4): 22-33.
[13]	王宁, 葛瑞芳, 苑春法, 等. 中文金融新闻中公司名的识别[J]. 中文信息学报, 2002, 16(2): 1-6.
	WANG N, GE R F, YUAN C F, et al. Company name iden-tification in Chinese financial domain[J]. Journal of Chinese Information Processing, 2002, 16(2): 1-6.
[14]	BIKEL D M, SCHWARTZ R, WEISCHEDEL R M. An algorithm that learns what's in a name[J]. Machine Lear-ning, 1999, 34: 211-231.
[15]	ZHOU G D, SU J. Named entity recognition using an HMM-based chunk tagger[C]// Proceedings of the 40th Annual Mee-ting on Association for Computational Linguistics, Philadel-phia, Jul 6-12, 2002. Stroudsburg: ACL, 2002: 473-480.
[16]	张华平, 刘群. 基于角色标注的中国人名自动识别研究[J]. 计算机学报, 2004, 27(1): 85-91.
	ZHANG H P, LIU Q. Automatic recognition of Chinese personal name based on role tagging[J]. Chinese Journal of Computers, 2004, 27(1): 85-91.
[17]	俞鸿魁, 张华平, 刘群, 等. 基于层叠隐马尔可夫模型的中文命名实体识别[J]. 通信学报, 2006, 27(2): 87-94.
	YU H K, ZHANG H P, LIU Q, et al. Chinese named entity identification using cascaded hidden Markov model[J]. Jour-nal on Communications, 2006, 27(2): 87-94.
[18]	BORTHWICK A, STERLING J, AGICHTEIN E, et al. NYU: description of the MENE named entity system as used in MUC-7[C]// Proceedings of the 7th Message Understanding Conference, Virginia, Apr 29-May 1, 1998. Stroudsburg: ACL, 1998: 1-7.
[19]	BENDER O, OCH F J, NEY H. Maximum entropy models for named entity recognition[C]// Proceedings of the 7th Con-ference on Natural Language Learning, Edmonton, May 31-Jun 1, 2003. Stroudsburg: ACL, 2003: 148-151.
[20]	周雅倩, 郭以昆, 黄萱菁, 等. 基于最大熵方法的中英文基本名词短语识别[J]. 计算机研究与发展, 2003, 40(3): 440-446.
	ZHOU Y Q, GUO Y K, HUANG X J, et al. Chinese and English base NP recognition based on a maximum entropy model[J]. Journal of Computer Research and Development, 2003, 40(3): 440-446.
[21]	张玥杰, 徐智婷, 薛向阳. 融合多特征的最大熵汉语命名实体识别模型[J]. 计算机研究与发展, 2008, 45(6): 1004-1010.
	ZHANG Y J, XU Z T, XUE X Y. Fusion of multiple features for Chinese named entity recognition based on maximum entropy model[J]. Journal of Computer Research and Development, 2008, 45(6): 1004-1010.
[22]	ISOZAKI H, KAZAWA H. Efficient support vector classi-fiers for named entity recognition[C]// Proceedings of the 19th International Conference on Computational Linguis-tics, Taipei, China, Aug 24-Sep 1, 2002. Stroudsburg: ACL, 2002: 1-7.
[23]	TAKEUCHI K, COLLIER N. Use of support vector ma-chines in extended named entity recognition[C]// Procee-dings of the 6th Conference on Natural Language Learning, Taipei, China, Aug 24-Sep 1, 2002. Stroudsburg: ACL, 2002: 184-190.
[24]	李丽双, 黄德根, 陈春荣, 等. 基于支持向量机的中文文本中地名识别[J]. 大连理工大学学报, 2007, 47(3): 433-438.
	LI L S, HUANG D G, CHEN C R, et al. Identification of location names from Chinese texts based on support vector machine[J]. Journal of Dalian University of Technology, 2007, 47(3): 433-438.
[25]	陈霄, 刘慧, 陈玉泉. 基于支持向量机方法的中文组织机构名的识别[J]. 计算机应用研究, 2008, 25(2): 362-364.
	CHEN X, LIU H, CHEN Y Q. Chinese organization names recognition based on SVM[J]. Application Research of Computers, 2008, 25(2): 362-364.
[26]	MCCALLUM A, LI W. Early results for named entity recognition with conditional random fields, feature induc-tion and web-enhanced lexicons[C]// Proceedings of the 7th Conference on Natural Language Learning, Edmonton, May 31-Jun 1, 2003. Stroudsburg: ACL, 2003: 188-191.
[27]	KRISHNAN V, MANNING C D. An effective two-stage model for exploiting non-local dependencies in named enti-ty recognition[C]// Proceedings of the 21st International Con-ference on Computational Linguistics and 44th Annual Meeting of the Association for Computational Linguistics, Sydney, Jul 17-21, 2006. Stroudsburg: ACL, 2006: 1121-1128.
[28]	冯元勇, 孙乐, 张大鲲, 等. 基于小规模尾字特征的中文命名实体识别研究[J]. 电子学报, 2008, 36(9): 1833-1838.
	FENG Y Y, SUN L, ZHANG D K, et al. Study on the Chinese named entity recognition using small scale character tail hints[J]. Acta Electronica Sinica, 2008, 36(9): 1833-1838.
[29]	燕杨, 文敦伟, 王云吉, 等. 基于层叠条件随机场的中文病历命名实体识别[J]. 吉林大学学报(工学版), 2014, 44(6): 1843-1848.
	YAN Y, WEN D W, WANG Y J, et al. Named entity recog-nition in Chinese medical records based on cascaded condi-tional random field[J]. Journal of Jilin University(Enginee-ring and Technology Edition), 2014, 44(6): 1843-1848.
[30]	TEIXEIRA J, SARMENTO L, OLIVEIRA E C. A boot-strapping approach for training a NER with conditional ran-dom fields[C]// LNCS 7026: Proceedings of the 15th Portu-guese Conference on Artificial Intelligence, Lisbon, Oct 10-13, 2011. Berlin, Heidelberg: Springer, 2011: 664-678.
[31]	THENMALAR S, BALAJI J, GEETHA T. Semi-supervised bootstrapping approach for named entity recognition[J]. arXiv:1511.06833, 2015.
[32]	黄诗琳, 郑小林, 陈德人. 针对产品命名实体识别的半监督学习方法[J]. 北京邮电大学学报, 2013, 36(2): 20-23.
	HUANG S L, ZHENG X L, CHEN D R. A semi-supervised learning method for product named entity recognition[J]. Journal of Beijing University of Posts and Telecommunica-tions, 2013, 36(2): 20-23.
[33]	LONG L Y, YAN J Z, FANG L Y, et al. The identification of Chinese named entity in the field of medicine based on bootstrapping method[C]// Proceedings of the 2014 Interna-tional Conference on Multisensor Fusion and Information Integration for Intelligent Systems, Beijing, Sep 28-29, 2014. Piscataway: IEEE, 2014: 1-6.
[34]	王路路, 艾山·吾买尔, 买合木提·买买提, 等. 基于CRF和半监督学习的维吾尔文命名实体识别[J]. 中文信息学报, 2018, 32(11): 16-26.
	WANG L L, AISHAN W, MAIHEMUTI M, et al. A semi-supervised approach to Uyghur named entity recognition based on CRF[J]. Journal of Chinese Information Proces-sing, 2018, 32(11): 16-26.
[35]	ETZIONI O, CAFARELLA M, DOWNEY D, et al. Unsu-pervised named-entity extraction from the web: an experi-mental study[J]. Artificial Intelligence, 2005, 165(1): 91-134. DOI URL
[36]	NADEAU D, TURNEY P D, MATWIN S. Unsupervised named-entity recognition: generating gazetteers and resol-ving ambiguity[C]// LNCS 4013: Proceedings of the 19th Conference of the Canadian Society for Computational Stu-dies of Intelligence, Quebec, Jun 7-9, 2006. Berlin, Heidel-berg: Springer, 2006: 266-277.
[37]	HAN X, KWOH C K, KIM J J. Clustering based active learning for biomedical named entity recognition[C]// Pro-ceedings of the 2016 International Joint Conference on Neural Networks, Vancouver, Jul 24-29, 2016. Piscataway: IEEE, 2016: 1253-1260.
[38]	DEVLIN J, CHANG M W, LEE K, et al. BERT: pre-training of deep bidirectional transformers for language unders-tanding[C]// Proceedings of the 2019 Conference of the North American Chapter of the Association for Computa-tional Linguistics: Human Language Technologies, Minnea-polis, Jun 2-7, 2019. Stroudsburg: ACL, 2019: 4171-4186.
[39]	COLLOBERT R, WESTON J, BOTTOU L, et al. Natural language processing (almost) from scratch[J]. Journal of Machine Learning Research, 2011, 12: 2493-2537.
[40]	YAO L, LIU H, LIU Y, et al. Biomedical named entity reco-gnition based on deep neutral network[J]. International Jour-nal of Hybrid Information Technology, 2015, 8(8): 279-288.
[41]	STRUBELL E, VERGA P, BELANGER D, et al. Fast and accurate entity recognition with iterated dilated convolu-tions[C]// Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, Copenhagen, Sep 9-11, 2017. Stroudsburg: ACL, 2017: 2670-2680.
[42]	WU Y H, JIANG M, LEI J B, et al. Named entity recog-nition in Chinese clinical text using deep neural network[C]// Proceedings of the 15th World Congress on Health and Biomedical Informatics, São Paulo, Aug 19-23, 2015: 624-628.
[43]	WU F Z, LIU J X, WU C H, et al. Neural Chinese named entity recognition via CNN-LSTM-CRF and joint training with word segmentation[C]// Proceedings of the World Wide Web Conference, San Francisco, May 13-17, 2019. New York: ACM, 2019: 3342-3348.
[44]	KONG J, ZHANG L X, JIANG M, et al. Incorporating multi-level CNN and attention mechanism for Chinese clini-cal named entity recognition[J]. Journal of Biomedical Informatics, 2021, 116:103737-103743. DOI URL
[45]	GUI T, MA R T, ZHANG Q, et al. CNN-based Chinese NER with lexicon rethinking[C]// Proceedings of the 28th Interna-tional Joint Conference on Artificial Intelligence, Macao, China, Aug 10-16, 2019: 4982-4988.
[46]	HUANG Z H, XU W, YU K. Bidirectional LSTM-CRF mo-dels for sequence tagging[J]. arXiv:1508.01991, 2015.
[47]	GREGORIC A Z, BACHRACH Y, COOPE S. Named en-tity recognition with parallel recurrent neural networks[C]// Proceedings of the 56th International Conference on Com-putational Linguistics, Melbourne, Jul 15-20, 2018. Strouds-burg: ACL, 2018: 69-74.
[48]	LI F, WANG Z, HUI S C, et al. Modularized interaction network for named entity recognition[C]// Proceedings of the 59th Annual Meeting of the Association for Compu-tational Linguistics and the 11th International Joint Confe-rence on Natural Language Processing, Aug 1-6, 2021. St-roudsburg: ACL, 2021: 200-209.
[49]	XU Y X, HUANG H Y, FENG C, et al. A supervised multi-head self-attention network for nested named entity recogni-tion[C]// Proceedings of the 35th AAAI Conference on Arti-ficial Intelligence, the 33rd Conference on Innovative App-lications of Artificial Intelligence, the 11th Symposium on Educational Advances in Artificial Intelligence, Feb 2-9, 2021. Menlo Park: AAAI, 2021: 14185-14193.
[50]	ZHANG Y, YANG J. Chinese NER using lattice LSTM[C]// Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics, Melbourne, Jul 15-20, 2018. Stroudsburg: ACL, 2018: 1554-1564.
[51]	LIU W, XU T G, XU Q H, et al. An encoding strategy based word-character LSTM for Chinese NER[C]// Procee-dings of the 2019 Conference of the North American Chap-ter of the Association for Computational Linguistics: Hu-man Language Technologies, Minneapolis, Jun 2-7, 2019. Stroudsburg: ACL, 2020: 5951-5960.
[52]	MA R T, PENG M L, ZHANG Q, et al. Simplify the usage of lexicon in Chinese NER[C]// Proceedings of the 58th Ann-ual Meeting of the Association for Computational Linguis-tics, Jul 5-10, 2020. Stroudsburg: ACL, 2019: 2379-2389.
[53]	CETOLI A, BRAGAGLIA S, O'HARNEY A D, et al. Graph convolutional networks for named entity recognition[C]// Proceedings of the 16th International Workshop on Tree-banks and Linguistic Theories, Prague, Jan 23-24, 2018. St-roudsburg: ACL, 2018: 37-45.
[54]	DING R X, XIE P J, ZHANG X Y, et al. A neural multi-digraph model for Chinese NER with gazetteers[C]// Pro-ceedings of the 57th Conference of the Association for Computational Linguistics, Florence, Jul 28-Aug 2, 2019. Stroudsburg: ACL, 2019: 1462-1467.
[55]	GUI T, ZOU Y C, ZHANG Q, et al. A lexicon-based graph neural network for Chinese NER[C]// Proceedings of the 2019 Conference on Empirical Methods in Natural Langu-age Processing and the 9th International Joint Conference on Natural Language Processing, Hongkong, China, Nov 3-7, 2019. Stroudsburg: ACL, 2019: 1040-1050.
[56]	TANG Z, WAN B, YANG L. Word-character graph convo-lution network for Chinese named entity recognition[J]. IEEE/ACM Transactions on Audio, Speech, and Language Processing, 2020, 28: 1520-1532. DOI URL
[57]	PENG M L, XING X Y, ZHANG Q, et al. Distantly super-vised named entity recognition using positive-unlabeled lear-ning[C]// Proceedings of the 57th Conference of the As-sociation for Computational Linguistics, Florence, Jul 28-Aug 2, 2019. Stroudsburg: ACL, 2019: 2409-2419.
[58]	YANG Y S, CHEN W L, LI Z H, et al. Distantly supervised NER with partial annotation learning and reinforcement learning[C]// Proceedings of the 27th International Confe-rence on Computational Linguistics, Florence, Jul 28-Aug 2, 2019. Stroudsburg: ACL, 2019: 2159-2169.
[59]	ZHANG H L, LIU L T, CHENG S Z, et al. Distant super-vision for Chinese temporal tagging[C]// Proceedings of the 3rd China Conference on Knowledge Graph and Semantic Computing, Tianjin, Aug 14-17, 2018. Cham: Springer, 2018: 14-27.
[60]	边俐菁. 基于深度学习和远程监督的产品实体识别及其领域迁移研究[D]. 上海: 上海财经大学, 2020.
	BIAN L J. Research on product entity recognition and domain migration based on deep learning and remote super-vision[D]. Shanghai: Shanghai University of Finance and Economics, 2020.
[61]	SOUZA F, NOGUEIRA R, LOTUFO R. Portuguese named en-tity recognition using BERT-CRF[J]. arXiv:1909.10649, 2019.
[62]	NASEEM U, KHUSHI M, REDDY V, et al. BioALBERT: a simple and effective pre-trained language model for bio-medical named entity recognition[C]// Proceedings of the 2021 International Joint Conference on Neural Networks, Shenzhen, Jul 18-22, 2021. Piscataway: IEEE, 2021: 1-7.
[63]	YANG Z W, MA J, CHEN H C, et al. HiTRANS: a hierar-chical transformer network for nested named entity recogni-tion[C]// Proceedings of the Findings of the Association for Computational Linguistics, Punta Cana, Nov 16-20, 2021. Stroudsburg: ACL, 2021: 124-132.
[64]	李妮, 关焕梅, 杨飘, 等. 基于BERT-IDCNN-CRF的中文命名实体识别方法[J]. 山东大学学报(理学版), 2020, 55(1): 102-109.
	LI N, GUAN H M, YANG P, et al. BERT-IDCNN-CRF for named entity recognition in Chinese[J]. Journal of Shan-dong University (Natural Science), 2020, 55(1): 102-109.
[65]	LI X Y, ZHANG H, ZHOU X H. Chinese clinical named entity recognition with variant neural structures based on BERT methods[J]. Journal of Biomedical Informatics, 2020, 107: 103422-103428. DOI URL
[66]	WU Y, HUANG J, XU C, et al. Research on named entity recognition of electronic medical records based on RoBERTa and radical-level feature[J]. Wireless Communications and Mobile Computing, 2021: 2489754.
[67]	YAO L G, HUANG H S, WANG K W, et al. Fine-grained mechanical Chinese named entity recognition based on ALBERT-AttBiLSTM-CRF and transfer learning[J]. Symme-try, 2020, 12(12): 1986-2006.
[68]	BROWN T, MANN B, RYDER N, et al. Language models are few-shot learners[C]// Advances in Neural Information Processing Systems 33, Dec 6-12, 2020: 1877-1901.
[69]	CUI L Y, WU Y, LIU J, et al. Template based named entity recognition using BART[C]// Proceedings of the Joint Con-ference of the 59th Annual Meeting of the Association for Computational Linguistics and the 10th International Joint Conference on Natural Language Processing, Bangkok, Aug 1-6, 2021. Stroudsburg: ACL, 2021: 1835-1845.
[70]	CHEN X, ZHANG N, LI L, et al. Lightner: a lightweight generative framework with prompt-guided attention for low-resource NER[J]. arXiv:2109.00720, 2021.
[71]	MA R, ZHOU X, GUI T, et al. Template-free prompt tuning for few-shot NER[J]. arXiv:2109.13532, 2021.
[72]	LIU A T, XIAO W, ZHU H, et al. QaNER: prompting ques-tion answering models for few-shot named entity recogni-tion[J]. arXiv:2203.01543, 2022.
[73]	金彦亮, 谢晋飞, 吴迪嘉. 基于分层标注的中文嵌套命名实体识别[J]. 上海大学学报(自然科学版), 2021, 27(3): 1-9.
	JIN Y L, XIE J F, WU D J. Chinese nested named entity recognition based on hierarchical tagging[J]. Journal of Shanghai University (Natural Science Edition), 2021, 27(3): 1-9.

数据集	时间	语言	语料来源	实体类型数量	URL
MUC-6 MUC-7	1996 1997	英文	新闻	共7类：人名、地名、机构名、日期、时间、货币和百分比	https://catalog.ldc.upenn.edu/LDC-2001T02
1998年人民日报数据集	1998	中文	人民日报	共3类：人名、地名和机构名	https://github.com/InsaneLife/Chine-seNLPCorpus/tree/master/NER/ren-MinRiBao
CoNLL2002- 2003	2002—2003	英文、德文	新闻	共4类：人名、地名、机构名和其他实体	https://www.clips.uantwerpen.be/conll2002/ner/
GENIA	2004	英文	生物学和临床文本	共36个细粒度实体类型	http://www.geniaproject.org/home
ACE2004- 2005	2004—2005	英文、阿拉伯文、中文	新闻、博客	共7类：人名、地名、机构名、地理政治、设施、交通工具和武器	https://www.ldc.upenn.edu/collaborations/past-projects/ace
MSRA	2006	中文	新闻	共3类：人名、地名和机构名	http://www.sighan.org/bakeoff2006/
OntoNotes5.0	2013	英文、阿拉伯文、中文	新闻、博客、宗教文本	共18类：人名、地名、机构名、地理政治、设施、产品、事件实体等	https://catalog.ldc.upenn.edu/LDC-2013T19
BosonNLP NER	2014	中文	网络文本	共6类：人名、地名、机构名、时间、公司名和产品实体	http://bosonnlp.com
NCBI Disease	2014	英文	PubMed摘要	共790个细粒度实体类型	https://www.ncbi.nlm.nih.gov/CBBre-search/Dogan/DISEASE/
BC5CDR	2015	英文	PubMed摘要	共3类：化学物质、疾病名和化学-疾病相互作用	http://bioc.sourceforge.net/
CCKS2017	2017	中文	电子病历	共5类：症状体征、检查检验、疾病诊断、治疗和身体部位	https://www.biendata.xyz/competition/CCKS2017_2/
Weibo NER	2018	中文	博客	共4类：人名、地名、机构名和地理-政治	https://github.com/hltcoe/golden-horse
Chinese resume	2018	中文	简历	共8类：人名、地名、机构名、国家、教育机构、职业、种族和职称	https://github.com/jiesutd/LatticeLSTM
CCKS2018	2018	中文	电子病历	共5类：解剖部位、症状描述、独立症状、药物和手术	https://www.biendata.xyz/competition/CCKS2018_1/
CCKS2019-2020	2019—2020	中文	电子病历	共6类：疾病和诊断、检查、检验、药物、手术以及解剖部位	https://www.biendata.xyz/competition/ccks_2020_2_1/
CLUENER2020	2020	中文	新闻	共10类：人名、地名、机构名、公司、政府等	https://github.com/CLUEbenchmark/CLUENER2020
Few-NERD	2021	英文	维基百科	共8个粗粒度、66个细粒度	https://ningding97.github.io/fewnerd/

数据集	时间	语言	语料来源	实体类型数量	URL
MUC-6 MUC-7	1996 1997	英文	新闻	共7类：人名、地名、机构名、日期、时间、货币和百分比	https://catalog.ldc.upenn.edu/LDC-2001T02
1998年人民日报数据集	1998	中文	人民日报	共3类：人名、地名和机构名	https://github.com/InsaneLife/Chine-seNLPCorpus/tree/master/NER/ren-MinRiBao
CoNLL2002- 2003	2002—2003	英文、德文	新闻	共4类：人名、地名、机构名和其他实体	https://www.clips.uantwerpen.be/conll2002/ner/
GENIA	2004	英文	生物学和临床文本	共36个细粒度实体类型	http://www.geniaproject.org/home
ACE2004- 2005	2004—2005	英文、阿拉伯文、中文	新闻、博客	共7类：人名、地名、机构名、地理政治、设施、交通工具和武器	https://www.ldc.upenn.edu/collaborations/past-projects/ace
MSRA	2006	中文	新闻	共3类：人名、地名和机构名	http://www.sighan.org/bakeoff2006/
OntoNotes5.0	2013	英文、阿拉伯文、中文	新闻、博客、宗教文本	共18类：人名、地名、机构名、地理政治、设施、产品、事件实体等	https://catalog.ldc.upenn.edu/LDC-2013T19
BosonNLP NER	2014	中文	网络文本	共6类：人名、地名、机构名、时间、公司名和产品实体	http://bosonnlp.com
NCBI Disease	2014	英文	PubMed摘要	共790个细粒度实体类型	https://www.ncbi.nlm.nih.gov/CBBre-search/Dogan/DISEASE/
BC5CDR	2015	英文	PubMed摘要	共3类：化学物质、疾病名和化学-疾病相互作用	http://bioc.sourceforge.net/
CCKS2017	2017	中文	电子病历	共5类：症状体征、检查检验、疾病诊断、治疗和身体部位	https://www.biendata.xyz/competition/CCKS2017_2/
Weibo NER	2018	中文	博客	共4类：人名、地名、机构名和地理-政治	https://github.com/hltcoe/golden-horse
Chinese resume	2018	中文	简历	共8类：人名、地名、机构名、国家、教育机构、职业、种族和职称	https://github.com/jiesutd/LatticeLSTM
CCKS2018	2018	中文	电子病历	共5类：解剖部位、症状描述、独立症状、药物和手术	https://www.biendata.xyz/competition/CCKS2018_1/
CCKS2019-2020	2019—2020	中文	电子病历	共6类：疾病和诊断、检查、检验、药物、手术以及解剖部位	https://www.biendata.xyz/competition/ccks_2020_2_1/
CLUENER2020	2020	中文	新闻	共10类：人名、地名、机构名、公司、政府等	https://github.com/CLUEbenchmark/CLUENER2020
Few-NERD	2021	英文	维基百科	共8个粗粒度、66个细粒度	https://ningding97.github.io/fewnerd/

方法	时间	支持语言	数据集	F1值/%	方法关键字
Krupka^[10]	1995	英文	MUC-6	96.42	人名和地名规则
Shaalan等^[11]	2009	阿拉伯文	ACE、政府提供以及网上数据构成的数据集	92.26	地名词典+规则
张小衡等^[12]	1997	中文	香港理工大学的三地现代汉语数据集	97.30	机构名规则
王宁等^[13]	2002	中文	互联网金融新闻构成的数据集	89.13	公司名规则

方法	时间	支持语言	数据集	F1值/%	方法关键字
Krupka^[10]	1995	英文	MUC-6	96.42	人名和地名规则
Shaalan等^[11]	2009	阿拉伯文	ACE、政府提供以及网上数据构成的数据集	92.26	地名词典+规则
张小衡等^[12]	1997	中文	香港理工大学的三地现代汉语数据集	97.30	机构名规则
王宁等^[13]	2002	中文	互联网金融新闻构成的数据集	89.13	公司名规则

模型	原理	优点	缺点	代表文献
HMM	对转移概率和表现概率直接建模,统计共现概率	时间复杂度低	准确率比MEM略低	[14⇓⇓-17]
MEM	对转移概率和表现概率建立联合概率,统计的是条件概率	准确率比HMM高	时间复杂度高	[18⇓⇓-21]
SVM	特征空间上的间隔最大的线性分类器	利用内积核函数代替向高维空间的非线性映射	大规模样本训练和多分类效果差	[22⇓⇓-25]
CRF	统计全局概率,不仅在局部归一化,考虑数据在全局的分布	考虑数据的全局分布,解决了标注偏置问题	时间复杂度高	[26⇓⇓-29]

命名实体识别方法研究综述

Review on Named Entity Recognition

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 11

参考文献 73

相关文章 15

编辑推荐

Metrics

方法	实现方法	域泛化能力	优点	缺点	改进方法
有监督学习	分类	最弱	充分利用先验知识,针对特定的域	需要大量的标注数据,可移植性差	增加特征,增加标记语料
半监督学习	分类	较强	需要少量的语料	需要大量的分析和后期处理	扩展模式,减少噪音
无监督学习	聚类	最强	不需要标注语料库,用于大规模未标注语料	需要提前确定聚类阈值,性能较低	扩展特征,改善聚类

方法	时间	支持语言	数据集	F1值/%	方法关键字
Borthwick等^[18]	1998	英文	MUC-7	92.05	MEM、知识库
Bikel等^[14]	1999	英文	MUC-6	94.92	HMM
Zhou等^[15]	2002	英文	MUC-6、MUC-7	96.60、94.10	HMM、特征扩充
Isozaki等^[22]	2002	英文	General数据集	90.03	SVM
Bender等^[19]	2003	英文	CoNLL2003	89.26	MEM
McCallum等^[26]	2003	英文	CoNLL2003	84.04	单层CRF
张华平等^[16]	2004	中文	人民日报数据集	人名95.40	HMM、角色标注
俞鸿魁等^[17]	2006	中文	人民日报数据集	均值91.20	层叠HMM
Krishnan等^[27]	2006	英文	CoNLL2003	87.24	双层CRF
Nadeau等^[36]	2006	英文	MUC-7	69.33	无监督学习、地名词典
张玥杰等^[21]	2008	中文	SIGHAN 2008	87.92	MEM、规则
李丽双等^[24]	2007	中文	人民日报数据集	90.12	SVM
陈霄等^[25]	2008	中文	人民日报数据集	84.18	SVM
冯元勇等^[28]	2008	中文	863简体NER评测数据集	88.76	单层CRF、尾字特征
Teixeira等^[30]	2011	英文	HAREM数据集	68.00	自举方法、CRF
黄诗琳等^[32]	2013	中文	新闻和网页文档构成的数据集	78.20	自举方法、规则和词典、CRF
燕杨等^[29]	2014	中文	临床医院65分电子病历数据集	97.02	层叠CRF
Thenmalar等^[31]	2015	英文	CoNLL2003	82.57	自举方法
Long等^[33]	2014	中文	医生相关的文本构成的数据集	均值94.75	自举方法
王路路等^[34]	2018	维吾尔文	新疆多语种信息技术实验室提供的数据集	87.43	半监督学习、CRF
Han等^[37]	2016	英文	BioCreative提供的数据集	81.40	聚类方法、主动学习

方法	代表模型	优点	缺点
有监督深度学习方法	CNN	数据处理并行化,对高维数据处理无压力	无法很好地提取序列信息
	RNN	解决了CNN无法记忆上下文信息的问题	会出现梯度消失和梯度爆炸现象
	LSTM	引入了输入门、输出门和遗忘门,解决了RNN长期依赖的问题,有效地学习长期依赖信息	梯度问题未完全解决
	GNN	利用图数据结构的数据处理模型,可以更高效地挖掘实体之间的联系	模型灵活性和扩展性差
远程监督深度学习方法	—	一定程度解决了需要大规模已标注数据问题	会产生不完全标注和噪音标注
Transformer方法	BERT类	采用掩码语言模型对双向的Transformer进行预训练,以生成深层的双向语言表征	需要大量GPU和训练数据
提示学习的方法	—	重新定义语言模型的输入,减少预训练模型和下游任务的差距,在低资源场景性能良好	计算复杂度高,需要人工设计提示

方法	时间	支持语言	数据集	F1值/%	方法关键字
Collobert等^[39]	2011	英文	RCV1	89.59	CNN-CRF、Gazetteer
Huang等^[46]	2015	英文	CoNLL2003	90.10	Bi-LSTM-CRF
Strubell等^[41]	2017	英文	CoNLL2003、OntoNotes5.0	90.65、86.84	ID-CNNs、RUG
Cetoli等^[53]	2018	英文	OntoNotes5.0	83.60	Bi-LSTM-GCN-CRF
Gregoric等^[47]	2018	英文	CoNLL2003	91.48	Multiple independent bidirectional LSTM units、Softmax
Zhang等^[50]	2018	中文	OntoNotes4.0、MSRA、Weibo NER、Chinese resume	73.88、93.18、58.79、94.46	Lattice-LSTM
Yang等^[56]	2020	中文	E-commerce-NER、NEWS NER	61.45、79.22	Distantly supervised、Bi-LSTM-MLPCRF
Gui等^[45]	2019	中文	OntoNotes4.0、MSRA、Weibo NER、Chinese resume	74.45、93.71、59.92、95.11	Lexicon rethinking、CNN
Liu等^[51]	2019	中文	OntoNotes4.0、MSRA、Weibo NER、Chinese resume	74.43、93.74、59.84、95.21	WC-LSTM
Ding等^[54]	2019	中文	OntoNotes4.0、MSRA、Weibo NER、E-commerce-NER	76.00、94.40、59.50、75.20	GGNN-LSTM-CRF
Gui等^[55]	2019	中文	OntoNotes4.0、MSRA、Weibo NER、Chinese resume	74.89、93.46、60.21、95.37	Lexicon-based、GCN
Wu等^[43]	2019	中文	Bakeoff-3、Bakeoff-4	89.42、90.18	CNN-LSTM-CRF、Joint training
Peng等^[57]	2019	英文	CoNLL2003、CoNLL2002、Twitter	82.94、75.85、59.36	Positive-unlabeled learning、Distantly supervised
Tang等^[56]	2020	中文	OntoNotes4.0、MSRA、Weibo NER、Chinese resume	75.87、94.40、63.63、95.53	Word-character representation、GCN
Ma等^[52]	2020	中文	OntoNotes4.0、MSRA、Weibo NER、Chinese resume	82.81、95.42、70.50、96.11	SoftLexicon、Bi-LSTM
李妮等^[64]	2020	中文	MSRA	94.42	BERT-IDCNN-CRF
Li等^[65]	2020	中文	CCKS2017、CCKS2018	91.60、89.56	BERT-Bi-LSTM-CRF
Kong等^[44]	2021	中文	CCKS2017、CCKS2019	90.49、85.13	Multi-Level CNN、Attention Mechanism
Li等^[48]	2021	英文	CoNLL2003	92.53	Modularized interaction network、RNN-BiLSTM-CRF
Xu等^[49]	2021	英文	ACE2004、ACE2005、GENIA	86.30、85.40、79.60	Multi-head self-attention、BERT
Naseem等^[62]	2021	英文	NCBI Disease、BC5CDR	97.18、97.78	BioALBERT
Yang等^[63]	2021	英文	ACE2004、ACE2005、GENIA	87.88、87.04、79.08	Hierarchical transformer
Wu等^[66]	2021	中文	CCKS2017、CCKS2019	93.26、82.87	RoBERTa、Radical-level feature
Cui等^[69]	2021	英文	CoNLL2003	92.55	BART、Multi-template
Chen等^[70]	2021	英文	CoNLL2003	93.90	Prompt-guided attention
Ma等^[71]	2021	英文	CoNLL2003、OntoNotes5.0	74.80、72.99	Template-free、Prompt tuning

[1]	吕晓琦, 纪科, 陈贞翔, 孙润元, 马坤, 邬俊, 李浥东. 结合注意力与循环神经网络的专家推荐算法[J]. 计算机科学与探索, 2022, 16(9): 2068-2077.
[2]	张祥平, 刘建勋. 基于深度学习的代码表征及其应用综述[J]. 计算机科学与探索, 2022, 16(9): 2011-2029.
[3]	任宁, 付岩, 吴艳霞, 梁鹏举, 韩希. 深度学习应用于目标检测中失衡问题研究综述[J]. 计算机科学与探索, 2022, 16(9): 1933-1953.
[4]	杨才东, 李承阳, 李忠博, 谢永强, 孙方伟, 齐锦. 深度学习的图像超分辨率重建技术综述[J]. 计算机科学与探索, 2022, 16(9): 1990-2010.
[5]	张立华, 张顺顺. 机器学习解构区域金融风险防控研究进展[J]. 计算机科学与探索, 2022, 16(9): 1969-1989.
[6]	曾凡智, 许露倩, 周燕, 周月霞, 廖俊玮. 面向智慧教育的知识追踪模型研究综述[J]. 计算机科学与探索, 2022, 16(8): 1742-1763.
[7]	安凤平, 李晓薇, 曹翔. 权重初始化-滑动窗口CNN的医学图像分类[J]. 计算机科学与探索, 2022, 16(8): 1885-1897.
[8]	刘艺, 李蒙蒙, 郑奇斌, 秦伟, 任小广. 视频目标跟踪算法综述[J]. 计算机科学与探索, 2022, 16(7): 1504-1515.
[9]	赵小明, 杨轶娇, 张石清. 面向深度学习的多模态情感识别研究进展[J]. 计算机科学与探索, 2022, 16(7): 1479-1503.
[10]	夏鸿斌, 肖奕飞, 刘渊. 融合自注意力机制的长文本生成对抗网络模型[J]. 计算机科学与探索, 2022, 16(7): 1603-1610.
[11]	韩毅, 乔林波, 李东升, 廖湘科. 知识增强型预训练语言模型综述[J]. 计算机科学与探索, 2022, 16(7): 1439-1461.
[12]	刘雅芬, 郑艺峰, 江铃燚, 李国和, 张文杰. 深度半监督学习中伪标签方法综述[J]. 计算机科学与探索, 2022, 16(6): 1279-1290.
[13]	孙方伟, 李承阳, 谢永强, 李忠博, 杨才东, 齐锦. 深度学习应用于遮挡目标检测算法综述[J]. 计算机科学与探索, 2022, 16(6): 1243-1259.
[14]	谢欣彤, 胡悦阳, 刘譞哲, 赵耀帅, 姜海鸥. 传播用户代表性特征学习的谣言检测方法[J]. 计算机科学与探索, 2022, 16(6): 1334-1342.
[15]	程卫月, 张雪琴, 林克正, 李骜. 融合全局与局部特征的深度卷积神经网络算法[J]. 计算机科学与探索, 2022, 16(5): 1146-1154.