基于条件随机场的中医临床病历命名实体抽取

doi:10.3969/j.issn.1000-3428.2014.09.062

计算机工程

基于条件随机场的中医临床病历命名实体抽取

刘　凯^1a,周雪忠^1a,1b ,于　剑^1a,1b ,张润顺²

(1. 北京交通大学a. 计算机与信息技术学院;b. 交通数据分析与挖掘北京市重点实验室,北京100044;2. 中国中医科学院广安门医院,北京100053)

收稿日期:2013-06-07 出版日期:2014-09-15 发布日期:2014-09-12
作者简介:刘　凯(1986 - ),男,硕士,主研方向:文本信息抽取;周雪忠(通讯作者),副教授;于　剑、张润顺,教授。
基金资助:
国家自然科学基金资助项目(61105055, 81230086);国家“863”计划基金资助项目(2012AA02A609);中央高校基本科研业务费专项基金资助项目(K13JB00140)。

Named Entity Extraction of Traditional Chinese Medicine Medical Records Based on Conditional Random Field

LIU Kai ^1a ,ZHOU Xue-zhong^1a,1b ,YU Jian ^1a,1b ,ZHANG Run-shun ²

(1a. School of Computer and Information Technology;1b. Beijing Key Lab of Traffic Data Analysis and Mining,Beijing Jiaotong University,Beijing 100044,China;2. Guang’anmen Hospital,China Academy of Chinese Medical Sciences,Beijing 100053,China)

Received:2013-06-07 Online:2014-09-15 Published:2014-09-12

摘要/Abstract

摘要： 中医临床病历是中医重要的科研数据资源,但目前临床病历仍以文本为主要表达形式,对病历数据深入分析的前提是进行结构化信息抽取,而命名实体抽取是其基础性步骤。针对中医临床病历的命名实体,如症状、疾病和诱因等的抽取问题,通过手工标注的413 份病历数据(以中文字为特征)与4 类特征模版,将条件随机场(CRF)、隐马尔科夫模型(HMM)和最大熵马尔科夫模型(MEMM)用于中医病历命名实体抽取的实验,并进行比较分析。结果表明,结合合适的特征模版,CRF 命名实体抽取方法取得了较好的性能,F1 值的症状达到0. 80,疾病名称达到0 . 74,诱因0. 74。与HMM 和MEMM 相比,CRF 有最高的准确率和召回率,是一种较为适用的中医临床病历命名实体抽取方法。

关键词: 中医临床病历, 命名实体抽取, 语料库标注系统, 条件随机场, 特征模板

Abstract: Traditional Chinese Medicine(TCM) medical records are the important data resources of the TCM medical research. The main form of them is still text now,and it is necessary to extract the structured information from the medical records,while named entity extraction is the basic step. It makes 413 copies of manually labeled medical records in Chinese text and four types of feature templates to study about the named entity extraction practice such as symptoms, diseases and incentives. It compares the results of TCM medical records named entity extraction by Conditional Random Field(CRF ), Hidden Markov Model ( HMM ) and Maximum Entropy Markov Model ( MEMM ). Combined with appropriate feature templates,CRF has well performance of F1:symptoms 0. 80,the name of the disease 0. 74,incentives 0. 74. Compared with HMM and MEMM,CRF has the highest precision and recall rate. This preliminary shows that CRF is an applicable method of the Chinese medical records named entity extraction

Key words: Traditional Chinese Medicine(TCM) medical records, named entity extraction, corpus annotation system, Conditional Random Field(CRF), feature template

中图分类号:

TP391

刘凯,周雪忠,于剑,张润顺. 基于条件随机场的中医临床病历命名实体抽取[J]. 计算机工程, doi: 10.3969/j.issn.1000-3428.2014.09.062.

LIU Kai,ZHOU Xue-zhong,YU Jian,ZHANG Run-shun. Named Entity Extraction of Traditional Chinese Medicine Medical Records Based on Conditional Random Field[J]. Computer Engineering, doi: 10.3969/j.issn.1000-3428.2014.09.062.

http://www.ecice06.com/CN/Y2014/V40/I9/312

参考文献

参考文献 [ 1 ]　周雪忠. 文本挖掘在中医药中的若干应用研究[D]. 杭州:浙江大学,2004. [ 2 ]　Zhou Xuezhong, Peng Yonghong, Liu Baoyan. Text Mining for Traditional Chinese Medical Knowledge Discovery: A Survey [ J ]. Journal of Biomedical Informatics,2010,43(4):650-660. [ 3 ]　Zhou Xuezhong, Liu Baoyan, Wang Yinghui, et al. Building Clinical Data Warehouse for Traditional Chinese Medicine Knowledge Discovery [C] / / Proc. of International Conference on BioMedical Engineering and Informatics. [S. l. ]:IEEE Press,2008:615-620. [ 4 ]　Zhou Xuezhong, Chen Shibo, Liu Baoyan, et al. Development of Traditional Chinese Medicine Clinical Data Warehouse for Medical Knowledge Discovery and Decision Support[J]. Artificial Intelligence in Medicine, 2010,48(2/ 3):139-152. [ 5 ]　Lafferty J D,McCallum A,Pereira F C N. Conditional Random Fields:Probabilistic Models for Segmenting and Labeling Sequence Data [ C ] / / Proc. of the 18th International Conference on Machine Learning. [S. l. ]: Morgan Kaufmann Publishers Inc. ,2001:282-289. [ 6 ]　熊　英. 中文自然语言理解中基于条件随机场理论的词法分析研究[D]. 上海:上海交通大学,2009. [ 7 ]　Franzén K,Eriksson G,Olsson F,et al. Protein Names and How to Find Them [ J ]. International Journal of medical Informatics,2002,67(1):49-61. [ 8 ]　Kim J D,Ohta T,Tsuruoka Y,et al. Introduction to the Bio-Entity Recognition Task at JNLPBA[C] / / Proc. of International Joint Workshop on Natural Language Processing in Biomedicine and Its Applications. [S. l. ]: IEEE Press,2004:70-75. [ 9 ]　宗成庆. 统计自然语言处理[M]. 北京:清华大学出版社,2008. [10]　Freitag D, McCallum A. Information Extraction with HMM Structures Learned by Stochastic Optimization [ C ] / / Proc. of the National Conference on Artificial Intelligence. [S. l. ]:AAAI Press,2000:584-589. [11]　McCallum A,Freitag D, Pereira F. Maximum Entropy Markov Models for Information Extraction and Segmentation [C] / / Proc. of the 17th International Conference on Machine Learning. Pittsburgh,USA:[s. n. ],2000: 591-598. [12]　McCallum A K. Mallet: A Machine Learning for Language Toolkit [ EB / OL ]. ( 2002-02-28 ). http:/ / mallet. cs. umass. edu. 编辑　顾逸斐

[1]	司逸晨, 管有庆. 基于Transformer编码器的中文命名实体识别模型[J]. 计算机工程, 2022, 48(7): 66-72.
[2]	李军怀, 陈苗苗, 王怀军, 崔颖安, 张爱华. 基于ALBERT-BGRU-CRF的中文命名实体识别方法[J]. 计算机工程, 2022, 48(6): 89-94,106.
[3]	崔丽平, 古丽拉·阿东别克, 王智悦. 基于有向图模型的旅游领域命名实体识别[J]. 计算机工程, 2022, 48(2): 306-313.
[4]	吕江海, 杜军平, 周南, 薛哲. 基于膨胀卷积迭代与注意力机制的实体名识别方法[J]. 计算机工程, 2021, 47(1): 58-65,71.
[5]	何阳宇, 晏雷, 易绵竹, 李宏欣. 融合CRF与规则的老挝语军事领域命名实体识别方法[J]. 计算机工程, 2020, 46(8): 297-304.
[6]	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): 40-45,52.
[7]	蔡凯, 李新福, 田学东. 基于视觉注意力分析的3D内容生成方法[J]. 计算机工程, 2020, 46(4): 266-272.
[8]	王仁武, 张文慧. 基于深度学习的隐性评价对象识别方法[J]. 计算机工程, 2019, 45(8): 315-320.
[9]	张杰妹, 杨词慧. 基于RV-FCN的CT肝脏影像自动分割算法[J]. 计算机工程, 2019, 45(7): 258-263.
[10]	张应成,杨洋,蒋瑞,全兵,张利君,任晓雷. 基于BiLSTM-CRF的商情实体识别模型[J]. 计算机工程, 2019, 45(5): 308-314.
[11]	谷兴龙, 谢珺, 靳红伟, 续欣莹. 基于词特征与语义特征的评价对象识别[J]. 计算机工程, 2019, 45(11): 218-224.
[12]	王俊强, 李建胜, 周华春, 张旭. 基于Deeplabv3+与CRF的遥感影像典型要素提取方法[J]. 计算机工程, 2019, 45(10): 260-265,271.
[13]	买买提阿依甫,吾守尔·斯拉木,帕丽旦·木合塔尔,杨文忠. 基于BiLSTM-CNN-CRF模型的维吾尔文命名实体识别[J]. 计算机工程, 2018, 44(8): 230-236.
[14]	李雁群,何云琪,钱龙华,周国栋. 基于维基百科的中文嵌套命名实体识别语料库自动构建[J]. 计算机工程, 2018, 44(11): 76-82.
[15]	陈炳丰,郝志峰,蔡瑞初,温雯,梁礼欣. 基于AWCRF模型的微博情感倾向分类方法[J]. 计算机工程, 2017, 43(7): 187-192.

选择文件类型/文献管理软件名称

选择包含的内容

基于条件随机场的中医临床病历命名实体抽取

Named Entity Extraction of Traditional Chinese Medicine Medical Records Based on Conditional Random Field

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

模态框（Modal）标题

选择文件类型/文献管理软件名称

选择包含的内容

基于条件随机场的中医临床病历命名实体抽取

Named Entity Extraction of Traditional Chinese Medicine Medical Records Based on Conditional Random Field

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价