中文电子病历中的时间关系识别

doi:10.11772/j.issn.1001-9081.2017082087

计算机应用 ›› 2018, Vol. 38 ›› Issue (3): 626-632.DOI: 10.11772/j.issn.1001-9081.2017082087

中文电子病历中的时间关系识别

孙健¹, 高大启¹, 阮彤¹, 殷亦超², 高炬², 王祺¹

1. 华东理工大学信息科学与工程学院, 上海 200237;
2. 上海中医药大学附属曙光医院, 上海 200021

收稿日期:2017-08-28 修回日期:2017-11-04 出版日期:2018-03-10 发布日期:2018-03-07
通讯作者: 王祺
作者简介:孙健(1993-),女,安徽六安人,硕士研究生,CCF会员,主要研究方向:信息抽取、知识图谱;高大启(1957-),男,湖北襄阳人,教授,博士,CCF会员,主要研究方向:机器嗅觉、智能理论、模式识别;阮彤(1973-),女,江苏扬州人,教授,博士,CCF会员,主要研究方向:信息抽取、知识图谱、数据质量评估;殷亦超(1983-),男,上海人,工程师,硕士,主要研究方向:医院信息化;高炬(1966-),男,上海人,主任医师,硕士,主要研究方向:医院行政管理、中西医结合治疗肝胆病;王祺(1993-),男,江苏苏州人,硕士研究生,CCF会员,主要研究方向:信息抽取、知识图谱、机器翻译。
基金资助:
国家863计划项目（2015AA020107）；国家科技支撑计划项目（2015BAH12F01-05）。

Recognition of temporal relation in Chinese electronic medical records

SUN Jian¹, GAO Daqi¹, RUAN Tong¹, YIN Yichao², GAO Ju², WANG Qi¹

1. School of Information Science and Engineering, East China University of Science and Technology, Shanghai 200237, China;
2. Shuguang Hospital Affiliated to Shanghai University of Traditional Chinese Medicine, Shanghai 200021, China

Received:2017-08-28 Revised:2017-11-04 Online:2018-03-10 Published:2018-03-07
Supported by:
This work is partially supported by the National High Technology Research and Development Program (863 Program) of China (2015AA020107), the National Key Technology Research and Development Program (2015BAH12F01-05).

摘要/Abstract

摘要： 中文电子病历中的时间关系包括句内时间关系和句间时间关系，其中，句内时间关系包括句内事件-事件的时间关系和句内事件-时间的时间关系，句间时间关系即是句间事件-事件的时间关系。把中文电子病历文本中的时间关系识别转化成实体对分类问题，针对句内时间关系的识别，制定了高准确率的启发式规则，并设计了基本特征、短语句法特征、依存特征和其他特征，训练分类器缓解句内时间关系的识别错误；针对句间时间关系的识别，在高准确率的启发式规则之外，设计了基本特征、短语句法特征和其他特征，训练分类器减少句间时间关系的识别错误。实验结果表明，当分别使用支持向量机（SVM）、SVM和随机森林（RF）算法时，所提方法在句内事件-事件、句内事件-时间和句间事件-事件的时间关系识别上的效果最好，其F₁值分别达到了84.0%、85.6%和63.5%。

关键词: 时间关系识别, 实体对分类, 句内时间关系, 句间时间关系, 语言特征

Abstract: The temporal relation or temporal links (denoted by the TLink tag) in Chinese electronic medical records includes temporal relations within a sentence (hereafter referred to as "within-sentence TLinks"), and between-sentence TLinks. Among them, within-sentence TLinks include event/event TLinks and event/time TLinks, and between-sentence TLinks include event/event TLinks. The recognition of temporal relation in Chinese electronic medical record was transformed into classification problem on entity pairs. Heuristic rules with high accuracy were developed and two different classifiers with basic features, phrase syntax, dependency features, and other features were trained to determine within-sentence TLinks. Apart from heuristic rules with high accuracy, basic features, phrase syntax, and other features were used to train the classifiers to determine between-sentence TLinks. The experimental results show that Support Vector Machine (SVM), SVM and Random Forest (RF) algorithms achieve the best performance of recognition on within-sentence event/event TLinks, within-sentence event/time TLinks and between-sentence event/event TLinks, with F₁-scores of 84.0%, 85.6% and 63.5% respectively.

Key words: temporal relation recognition, entity pair classification, within-sentence temporal relation, between-sentence temporal relation, linguistic feature

中图分类号:

TP391.1

孙健, 高大启, 阮彤, 殷亦超, 高炬, 王祺. 中文电子病历中的时间关系识别[J]. 计算机应用, 2018, 38(3): 626-632.

SUN Jian, GAO Daqi, RUAN Tong, YIN Yichao, GAO Ju, WANG Qi. Recognition of temporal relation in Chinese electronic medical records[J]. Journal of Computer Applications, 2018, 38(3): 626-632.

参考文献

[1] SUN W, RUMSHISKY A, UZUNER O. Temporal reasoning over clinical text:the state of the art[J]. Journal of the American Medical Informatics Association, 2013, 20(5):814-819.
[2] SUN W, RUMSHISKY A, UZUNER O. Evaluating temporal relations in clinical text:2012 i2b2 Challenge[J]. Journal of the American Medical Informatics Association, 2013, 20(5):806-813.
[3] 王昀,苑春法.基于转换的时间-事件关系映射[J].中文信息学报,2004,18(4):24-31.(WANG Y, YUAN C F. A time-event mapping method based transformation[J]. Journal of Chinese Information Processing, 2004, 18(4):24-31.)
[4] 林静,苑春法.汉语时间关系抽取与计算[J].中文信息学报,2009,23(5):62-67.(LIN J, YUAN C F. Extraction and computation of Chinese temporal relation[J]. Journal of Chinese Information Processing, 2009, 23(5):62-67.)
[5] 王风娥,谭红叶,钱揖丽.基于最大熵的句内时间关系识别[J].计算机工程,2012,38(4):37-39.(WANG F E, TAN H Y, QIAN Y L. Recognition of temporal relation in one sentence based on maximum entropy[J]. Computer Engineering, 2012, 38(4):37-39.)
[6] 刘莉.中文时间事件关系识别的方法研究[D].重庆:重庆大学,2012:33-35.(LIU L. Research on automatic identification of temporal relations between Chinese time expressions and events[D]. Chongqing:Chongqing University, 2012:33-35.)
[7] D'SOUZA J, NG V. Classifying temporal relations in clinical data:a hybrid, knowledge-rich approach[J]. Journal of Biomedical Informatics, 2013, 46:S29-S39.
[8] NIKFARJAM A, EMADZADEH E, GONZALEZ G. Towards generating a patient's timeline:extracting temporal relationships from clinical notes[J]. Journal of Biomedical Informatics, 2013, 46:S40-S47.
[9] CHENG Y, ANICK P, HONG P, et al. Temporal relation discovery between events and temporal expressions identified in clinical narrative[J]. Journal of Biomedical Informatics, 2013, 46:S48-S53.
[10] DO H W, JEONG Y S. Temporal relation classification with deep neural network[C]//Proceedings of the 2016 International Conference on Big Data and Smart Computing. Washington, DC:IEEE Computer Society, 2016:454-457.
[11] CHENG F, MIYAO Y. Classifying temporal relations by bidirectional LSTM over dependency paths[C]//Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Short Papers). Stroudsburg, PA:Association for Computational Linguistics, 2017, 2:1-6.
[12] DLIGACH D, MILLER T, CHEN L, et al. Neural temporal relation extraction[C]//Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics. Stroudsburg, PA:Association for Computational Linguistics, 2017,2:746-751.
[13] XU D, ZHANG M, ZHAO T, et al. Data-driven information extraction from Chinese electronic medical records[J]. PLOS ONE, 2015, 10(8):e0136270.
[14] HANKCS. HanLP[EB/OL].[2014-12-23]. https://github.com/hankcs/HanLP.
[15] STOLCKE A. An efficient probabilistic context-free parsing algorithm that computes prefix probabilities[J]. Computational Linguistics, 1994, 21(2):165-201.
[16] DAN K, MANNING C D. Fast exact inference with a factored model for natural language parsing[EB/OL].[2017-03-03]. http://www.ling.helsinki.fi/kit/2008s/clt350/docs/KleinManning-lexpar03.pdf.
[17] HOLMES G, DONKIN A, WITTEN I H. WEKA:a machine learning workbench[EB/OL].[2017-03-03]. https://researchcommons.waikato.ac.nz/bitstream/handle/10289/1138/uow-cs-wp-1994-09.pdf;jsessionid=0E603B1F904CF7546CF087199832DD44?sequence=1.

中文电子病历中的时间关系识别

Recognition of temporal relation in Chinese electronic medical records

PDF

可视化

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

[1]	张阳, 王小宁. 基于Word2Vec词嵌入和高维生物基因选择遗传算法的文本特征选择方法[J]. 《计算机应用》唯一官方网站, 2021, 41(11): 3151-3155.
[2]	李灿, 杨雅婷, 马玉鹏, 董瑞. 基于语种相似性挖掘的神经机器翻译语料库扩充方法[J]. 《计算机应用》唯一官方网站, 2021, 41(11): 3145-3150.
[3]	李志超, 吐尔地·托合提, 艾斯卡尔·艾木都拉. 基于动态注意力和多角度匹配的答案选择模型[J]. 《计算机应用》唯一官方网站, 2021, 41(11): 3156-3163.
[4]	殷雨昌王洪元陈莉冯尊登肖宇. 基于单标注样本的多损失学习与联合度量视频行人重识别[J]. 计算机应用, 0, (): 0-0.
[5]	左亚尧陈致然洪嘉伟陈坤. 融合多语义特征的命名实体识别方法[J]. 计算机应用, 0, (): 0-0.
[6]	袁景凌, 丁远远, 潘东行, 李琳. 基于时序和上下文特征的中文隐式情感分类模型[J]. 计算机应用, 2021, 41(10): 2820-2828.
[7]	杨书新, 张楠. 融合情感词典与上下文语言模型的文本情感分析[J]. 计算机应用, 2021, 41(10): 2829-2834.
[8]	杨璐, 何明祥. 基于门控机制和卷积神经网络的中文文本情感分析模型[J]. 计算机应用, 2021, 41(10): 2842-2848.
[9]	董永峰, 刘超, 王利琴, 李英双. 融合多跳关系路径信息的关系推理方法[J]. 计算机应用, 2021, 41(10): 2799-2805.
[10]	吴赛赛, 梁晓贺, 谢能付, 周爱莲, 郝心宁. 面向领域实体关系联合抽取的标注方法[J]. 计算机应用, 2021, 41(10): 2858-2863.
[11]	胡婕胡燕刘梦赤张龑. 基于知识库实体增强BERT模型的中文命名实体识别[J]. 计算机应用, 0, (): 0-0.
[12]	郝志刚秦丽李国亮. 基于多属性综合评价的食品安全标准引用网络重要节点发现方法[J]. 计算机应用, 0, (): 0-0.
[13]	丁行硕李翔谢乾. 基于标签分层延深建模的企业画像构建方法[J]. 计算机应用, 0, (): 0-0.
[14]	刘子辰, 李小娟, 韦伟. 基于循环神经网络的专利价格自动评估[J]. 计算机应用, 2021, 41(9): 2532-2538.
[15]	余敦辉, 万鹏, 王社. 基于企业知识图谱构建的实体关联查询系统[J]. 计算机应用, 2021, 41(9): 2510-2516.