计算机科学 ›› 2018, Vol. 45 ›› Issue (2): 261-268.doi: 10.11896/j.issn.1002-137X.2018.02.045
冯艳红,于红,孙庚,孙娟娟
FENG Yan-hong, YU Hong, SUN Geng and SUN Juan-juan
摘要： 传统的命名实体识别方法直接依靠大量的人工特征和专门的领域知识,解决了监督学习语料不足的问题,但设计人工特征和获取领域知识的代价昂贵。针对该问题,提出一种基于BLSTM(Bidirectional Long Short-Term Memory)的神经网络结构的命名实体识别方法。该方法不再直接依赖于人工特征和领域知识,而是利用基于上下文的词向量和基于字的词向量,前者表达命名实体的上下文信息,后者表达构成命名实体的前缀、后缀和领域信息；同时,利用标注序列中标签之间的相关性对BLSTM的代价函数进行约束,并将领域知识嵌入模型的代价函数中,进一步增强模型的识别能力。实验表明,所提方法的识别效果优于传统方法。
| [1] SUN L P,GUO G,TANG W W,et al.Enterprise Abbreviation Prediction Based on Constitution Pattern and Conditional Random Field[J].Journal of Computer Applications,2016,6(2):449-454.(in Chinese) 孙丽萍,过弋,唐文武,等.基于构成模式和条件随机场的企业简称预测[J].计算机应用,2016,6(2):449-454. [2] DUAN H Z,ZHENG Y.A Study on Features of the CRFs-based Chinese Named Entity Recognition[J].International Journal of Advanced Intelligence Paradigms,2011,3(2):287-294. [3] HUANG D G,LI Z Z,WAN R.Chinese Organization Name Reco-gnition Using Cascaded Model Based on SVM and CRF[J].Journal of Dalian University of Technology,2010,0(5):782-787.(in Chinese) 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,2010,0(5):782-787. [4] FENG Y H,YU H,SUN G,et al.Domain-specific Terminology Recognition Method Based on Word Embedding and CRF[J].Journal of Computer Applications,2016,6(11):3146-3151.(in Chinese) 冯艳红,于红,孙庚,等.基于词向量和CRF的领域术语识别方法[J].计算机应用,2016,6(11):3146-3151. [5] FENG Y T,ZHANG H J,HAO W N.Named Entity Recognition for Military Text[J].Computer Science,2015,2(7):15-18,7.(in Chinese) 冯蕴天,张宏军,郝文宁.面向军事文本的命名实体识别[J].计算机科学,2015,2(7):15-18,47. [6] BENGIO Y,SCHWENK H,WENCAL J S,et al.A NeuralProbabilistic Language Model[J].Springer Berlin Heidelberg,2006,3(6):1137-1155. [7] MNIH A,HINTON G E.A Scalable Hierarchical DistributedLanguage Model[C]∥NIPS2008:Advances in Neural Information Processing Systems 21.Vancouver,British Columbia,Canada:Curran Associates,Inc.,2008:1081-1088. [8] MIKOLOV T,CHEN K,CORRADO G,et al.Efficient Estimation of Word Representations in Vector Space[J].Neural Computation,2014,4:1771-1800. [9] LI L S,HE H L,LIU S S,et al.Research of Word Representations on Biomedical Named Entity Recognition[J].Journal of Chinese Computer System,2016(2):302-307. [10] WANG G Y,CAI Y Q,GE F J.Using Hybrid Neural Network to Address Chinese Named Entity Recognition[C]∥IEEE,International Conference on Cloud Computing and Intelligence Systems.IEEE,2014:433-438. [11] WANG G Y.Research of Chinese Named Entity RecognitionBased on Deep Learning[D].Beijing:Beijing University of Technology,2015:33-38.(in Chinese) 王国昱.基于深度学习的中文命名实体识别研究[D].北京:北京工业大学,2015:33-38. [12] COLLOBERT R,WESTON J,BOTTOU L,et al.Natural Language Processing (almost) from Scratch[J].The Journal of Machine Learning Research,2011,2(1):2493-2537. [13] HOCHREITER S,SCHMIDHUBER J.Long Short-term Me-mory[J].Neural computation,1997,9(8):1735-1780. [14] GRAVES A,MOHAMED A R,HINTON G.Speech Recognition with Deep Recurrent Neural Networks[C]∥IEEE International Conference on Acoustics,Speech and Signal Processing,2013.Vancouver,BC,Canada,2013:6645-6649. [15] SCHUSTER M,PALIWAL K K.Bidirectional Recurrent Neural Networks[J].IEEE Transactions on Signal Processing,1997,5(11):2673-2681. [16] GRAVES A,SCHMIDHUBER J.Framewise Phoneme Classification with Bidirectional LSTM and Other Neural Network Architectures[J].Neural Networks,2005,8(5):602-610. [17] NIELSEN M A.Neural Networks and Deep Learning[EB/OL].(2015)[2017-02-14].http:// Neural networks and deep lear-ning.com/chapter3.html. [18] MIKOLOV T,SUTSKEVER I,CHEN K,et al.Distributed Re-presentations of Words and Phrases and Their Compositiona-lity[C]∥NIPS2013:Advances in Neural Information Processing Systems 26.Lake Tahoe,Nevada,USA:Curran Associates,Inc.2013:3111-3119. [19] WANG L,LU′IS T,MARUJO L,et al.2015b.Finding Function in Form:Compositional Character Models for Open Vocabulary Word Representation[C]∥Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP).2015. [20] LAI S W.Word and Document Embedding Based on NeuralNetwork Approaches[D].Beijing:University of Chinese Academy of Sciences,2016:27-39.(in Chinese) 来斯惟.基于神经网络的词和文档语义向量表示方法研究[D].北京:中国科学院大学,2016:27-39. [21] 搜狗全网新闻数据(SogouCA)[EB/OL].[2017-02-14].http://www.sogou.com/labs/dl/ca.html. [22] 搜狗输入法词库[EB/OL].[2017-02-14].http://pinyin.sogou.com/dict. [23] ZAREMBA W,SUTSKEVER I,VINYALS O.Recurrent Neural Network Regularization[EB/OL].(2015-2)[2016-09].http://arxiv.org/abs/1409.2329v5. |
| No related articles found! |
|
||
首页