融合字词模型的中文命名实体识别研究

摘要
参考文献
相关文章 (15)

全文: PDF (2048 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要命名实体识别(NER)是自然语言处理中一项非常重要的基础任务。传统的机器学习方法在处理该任务时,主要依赖于人们的专业领域知识和人工提取的特征。为了在不需要人工特征的条件下获得较好的结果,该文提出了一种融合字词BiLSTM模型的命名实体识别方法。首先分别用BiLSTM-CRF训练得到基于字的模型Char-NER和基于词的模型Word-NER,然后将两个模型得到的分值向量进行运算和拼接,将拼接后的向量作为特征送入SVM进行训练,使用SVM对Char-NER和Word-NER进行模型融合。实验结果表明,该方法在不需要人工特征的条件下,在1998年《人民日报》语料和MSRA语料上对人名、地名、机构名识别的F值分别达到了94.04%、92.15%、87.05%和91.73%、93.20%、83.15%。

	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
	殷章志
	李欣子
	黄德根
	李玖一

关键词 ：命名实体识别, BiLSTM-CRF, 模型融合, SVM

Abstract：Named Entity Recognition(NER) plays an important role in Natural Language Processing. In order to obtain better results without manual features, this paper proposes an NER method based on an ensemble model of BiLSTM. Firstly, we apply the BiLSTM-CRF training on the data, obtaining the character-based model Char-NER and the word-based model Word-NER respectively. Then the score vectors obtained by the two models are merged as the input to the SVM model. The experimental results show that this method achieves 94.04%, 92.15%, 87.05% and 91.73%, 93.20%, 83.15% F-Scores of name, location and organization on the 1998 people's daily and MSRA corpus respectively without hand-crafted features.

Key words： named entity recognition BiLSTM-CRF model ensemble SVM

收稿日期: 2018-12-25

中图分类号:

TP391

基金资助:国家自然科学基金(61672127)

作者简介: 殷章志(1993—),硕士研究生,主要研究领域为自然语言处理。E-mail:yinzhangzhi@mail.dlut.edu.cn
李欣子(1987—),硕士研究生,主要研究领域为自然语言处理。E-mail:xinzi_sunny@qq.com
黄德根(1965—),通信作者,教授,博士生导师,主要研究领域为自然语言处理、机器翻译、人工智能。E-mail:huangdg@dlut.edu.cn

链接本文:

http://jcip.cipsc.org.cn/CN/Y2019/V33/I11/95

[1] 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,(2): 87-94.
[2] Chieu H L,Ng H T.Named entity recognition: A maximum entropy approach using global information[C]//Proceedings of the 19th International Conference on Computational Linguistics,2002(1): 1-7.
[3] 黄德根,李泽中,万如. 基于SVM和CRF的双层模型中文机构名识别[J]. 大连理工大学学报,2010,50(5): 782-787.
[4] Aaron L F H,Derek F W,Lidia S C. Chinese named entity recognition with conditional random fields in the light of chinese characteristics[M]. Warsaw: Springer,2013.
[5] Collobert R,Weston J,Bottou L,et al.Natural language processing (almost) from scratch[J]. The Journal of Machine Learning Research,2011(12): 2493-2537.
[6] Huang Z,Xu W,Yu K.Bidirectional LSTM-CRF models for sequence tagging[J].arXiv preprint arXiv: 1508.01991,2015.
[7] Dong C,Zhang J,Zong C,et al. Character-based LSTM-CRF with radical-level features for chinese named entity recognition[C]//Proceedings of the International Conference on Computer Processing of Oriental Languages. Springer International Publishing,2016: 239-250.
[8] Levow G A. The third international chinese language processing bakeoff: Word segmentation and named entity recognition[C]//Proceedings of the Fifth SIGHAN Workshop on Chinese Language Processing,Sydney: Association for Computational Linguistics,2006: 108-117.
[9] Breiman L. Bagging predictors[J]. Machine Learning,1996,24(2),123-140.
[10] 潘泉,王增福,梁彦,等. 信息融合理论的基本方法与进展(Ⅱ)[J]. 控制理论与应用,2012,29(10): 1233-1244.
[11] 张佳宝. 基于条件随机场的中文命名实体识别研究[D]. 北京: 国防科学技术大学硕士学位论,2010.
[12] 冯蕴天,张宏军,郝文宁,等. 基于深度信念网络的命名实体识别[J]. 计算机科学,2016,43(04): 224-230.
[13] Zhou J,He L,Dai X,et al. Chinese named entity recognition with a multi-phase model[C]//Proceedings of the Fifth SIGHAN Workshop on Chinese Language Processing. Stroudsburg: Association for Computational Linguistics,2006: 213-216.
[14] Zhou J,Qu W,Zhang F. Chinese named entity recognition via joint identification and categorization[J]. Chinese Journal of Electronics,2013,22(2): 225-230.
[15] 王蕾,谢云,周俊生,等. 基于神经网络的片段级中文命名实体识别[J].中文信息学报,2018,32(03): 84-90,100.
[16] Hochreiter S,Schmidhuber J. Long short-term memory[J]. Neural Computation,1997,9(8): 1735-1780.
[17] Graves A, Mohamed A, Hinton G. Speech recognition with deep recurrent neural networks[C]//Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing IEEE,2013: 6645-6649.