[1]陈克金,叶善力.基于ERNIE与多特征融合的中文命名实体识别[J].浙江科技学院学报,2023,(05):421-429456.[doi:10.3969/j.issn.1671-8798.2023.05.008]
　CHEN Kejin,YE Shanli.Chinese named entity recognition based on ERNIE and multi-feature fusion[J].,2023,(05):421-429456.[doi:10.3969/j.issn.1671-8798.2023.05.008]
点击复制

基于ERNIE与多特征融合的中文命名实体识别(/HTML)
分享到：

《浙江科技学院学报》[ISSN:1001-3733/CN:61-1062/R]

卷:
期数:
2023年05期
页码:
421-429456
栏目:
出版日期:
2023-10-31

文章信息/Info

Title:
Chinese named entity recognition based on ERNIE and multi-feature fusion
文章编号:
1671-8798(2023)05-0421-09
作者:
陈克金; 叶善力
(浙江科技学院 理学院，杭州 310023)
Author(s):
CHEN Kejin;  YE Shanli
(School of Science, Zhejiang University of Science and Technology, Hangzhou 310023, Zhejiang, China)
关键词:
命名实体识别; ERNIE; 双向长短期记忆网络; 膨胀卷积神经网络; 注意力机制
分类号:
TP391.43
DOI:
10.3969/j.issn.1671-8798.2023.05.008
文献标志码:
A
摘要:
【目的】在中文命名实体识别中，传统命名实体识别方法中词向量只能将其映射为单一向量，无法表征一词多义，在特征提取过程中易忽略局部特征。针对以上问题，提出一种基于知识增强语义表示(enhanced reprsentation through knowledge integration，ERNIE)与多特征融合的实体识别方法。【方法】首先，通过预训练模型ERNIE获得词向量; 然后将词向量并行输入双向长短时记忆网络(bidirectional long short-term memory network，BiLSTM)与膨胀卷积神经网络(iterated dilated convolutional neural network，IDCNN)中提取特征，再将输出特征向量进行融合; 最后通过条件随机场(conditional random field，CRF)解码获取最佳序列。【结果】本研究所提出的模型优于其他传统模型，在微软亚洲研究院(Microsoft Research Asia，MSRA)数据集上的F1值达到了95.18%，相比基准模型BiLSTM-CRF F1值提高了8.86百分点，相比ERNIE-BiLSTM-CRF模型F1值提高了1.34百分点。此外，在ERNIE-BiLSTM-IDCNN-CRF中引入注意力机制后F1值仅提升了0.07百分点，可见引入注意力机制对本研究所提出的模型之识别效果提升有限。【结论】本研究所提出的模型有效地提升了中文数据集上的实体识别性能，可为自然语言处理的命名实体识别研究提供参考。

参考文献/References:

[1] LI J, SUN A X, HAN J L, et al. A survey on deep learning for named entity recognition[J].IEEE Transactions on Knowledge and Data Engineering,2020,34(1):50.
[2] OTTER D W, MEDINA J R, KALITA J K. A survey of the usages of deep learning for natural language processing[J].IEEE Transactions on Neural Networks and Learning Systems,2021,32(2):604.
[3] 赵华茗,钱力,余丽.依存句法特征的科研命名实体识别算法[J].图书情报工作,2020,64(11):108.
[4] 葛君伟,涂兆昊,方义秋.基于融合CNN和Transformer的分离结构机器翻译模型[J].计算机应用研究,2022,39(2):432.
[5] 冯静,李正武,张登云,等.基于隐马尔可夫模型的桥梁检测文本命名实体识别[J].交通世界,2020(8):32.
[6] 邵诗韵,周宇,杨蕾,等.基于条件随机场的电力工程标书文本实体识别方法[J].计算机与现代化,2020(12):72.
[7] HAMMERTON J. Named entity recognition with long short-term memory[C]//Proceedings of the Seventh Conference on Natural Language Learning at HLT-NAACL 2003. Edmonton:CoNLL,2003:172.
[8] LAMPLE G, BALLESTEROS M, SUBRAMANIAN S, et al. Neural architectures for named entity recognition[C]//Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics:Human Language Technologies. San Diego:NAACL,2016:260.
[9] TANG P, YANG P, SHI Y, et al. Recognizing Chinese judicial named entity using BiLSTM-CRF[J].Journal of Physics:Conference Series,2020,1592(1):12040.
[10] 马欢欢,孔繁之,高建强.中文电子病历命名实体识别方法研究[J].医学信息学杂志,2020,41(4):24.
[11] LUO L, YANG Z, YANG P, et al. An attention-based BiLSTM-CRF approach to document-level chemical named entity recognition[J].Bioinformatics,2018,34(8):1381.
[12] STRUBELL E, VERGA P, BELANGER D, et al. Fast and accurate entity recognition with iterated dilated convolutions[C]//Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing. Stroudsburg:EMNLP,2017:2670.
[13] 梁文桐,朱艳辉,詹飞,等.基于BERT的医疗电子病历命名实体识别[J].湖南工业大学学报,2020,34(4):54.
[14] MIKOLOV T, CHEN K, CORRADO G, et al. Efficient estimation of word representations in vector space[C]//Proceedings of International Conference on Learning Representations. Scottsdale:ICLR,2013:1.
[15] PETERS M, NEUMANN M, IYYER M, et al. Deep contextualized word representations[C]//The North American Chapter of the Association for Computational Linguistics. Louisiana:NAACL,2018:2227.
[16] RADFORD A, NARASIMHAN K, SALIMANS T, et al. Improving language understanding bygenerative pre-training[EB/OL].(2018-06-14)[2022-08-19].
[17] DEVLIN J, CHANG M W, LEE K, et al. Bert:pre-training of deep bidirectional transformers for language understanding[EB/OL].(2018-10-11)[2022-08-19].
[18] SUN Y, WANG S, LI Y, et al. Ernie:enhanced representation through knowledge integration[EB/OL].(2019-04-19)[2022-08-19].
[19] QIU J, ZHOU Y, WANG Q, et al. Chinese clinical named entity recognition using residual dilated convolutional neural network with conditional random field[J].IEEE Transactions on Nanobioscience,2019,18(3):306.
[20] 焦凯楠,李欣,朱容辰.中文领域命名实体识别综述[J].计算机工程与应用,2021,57(16):1.
[21] 刘晓俊,辜丽川,史先章.基于Bi-LSTM和注意力机制的命名实体识别[J].洛阳理工学院学报(自然科学版),2019,29(1):65.
[22] DONG C, ZHANG J, ZONG C, et al. Character-based LSTM-CRF with radical-level features for Chinese named entity recognition[C]//Natural lanquage understanding and intelligent applications:5th CCF conference on natural lanquage processing and Chinese computing. Kunming:NLPCC,2016:239.
[23] ZHANG Y, YANG J. Chinese NER using lattice LSTM[C]//Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics. Melbourne:ACL,2018:1554.
[24] JIA Y Z, XU X B. Chinese named entity recognition based on CNN-BiLSTM-CRF[C]//International Conference on Software Engineering and Service Science. Peking:IEEE,2018:1.
[25] 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.
[26] 杨飘,董文永.基于BERT嵌入的中文命名实体识别方法[J].计算机工程,2020,46(4):40.

相似文献/References:

[1]毕云杉,钱亚冠,张超华,等.基于ERNIE模型的中文文本分类研究[J].浙江科技学院学报,2021,(06):461.
　BI Yunshan,QIAN Yaguan,ZHANG Chaohua,et al.Research on Chinese text classification based on ERNIE model[J].,2021,(05):461.
[2]张杰,黄杰,万健.基于半监督学习的中文电子病历命名实体识别[J].浙江科技学院学报,2022,(06):502.
　ZHANG Jie,HUANG Jie,WAN Jian.On named entity recognition for Chinese electronic medical record based on semisupervised learning[J].,2022,(05):502.

备注/Memo

备注/Memo:
收稿日期:2022-08-19
基金项目:国家自然科学基金项目(11671357)
通信作者:叶善力(1967— )，男，福建省福州人，教授，博士，主要从事复函数空间、时间序列分析等研究。E-mail:slye@zust.edu.cn。
更新日期/Last Update: 2023-10-31