多维字符特征表示的铁路设备事故信息抽取方法

doi:10.16265/j.cnki.issn1003-3033.2022.06.2732

中国安全科学学报 ›› 2022, Vol. 32 ›› Issue (6): 109-114.doi: 10.16265/j.cnki.issn1003-3033.2022.06.2732

多维字符特征表示的铁路设备事故信息抽取方法

张鹏翔()

中国铁道科学研究院集团有限公司标准计量研究所,北京 100081

收稿日期:2022-01-14 修回日期:2022-04-11 出版日期:2022-06-28 发布日期:2022-12-28
作者简介:
张鹏翔 (1988—),男,甘肃通渭人,硕士,工程师,主要从事铁路运输方面的工作。E-mail: 790416708@qq.com。

Information extraction method for railway equipment accidents based on multi-dimensional character feature representation

ZHANG Pengxiang()

Standards & Metrology Research Institute, China Academy of Railway Sciences Corporation Limited, Beijing 100081, China

Received:2022-01-14 Revised:2022-04-11 Online:2022-06-28 Published:2022-12-28

摘要/Abstract

摘要：

为解决铁路设备事故调查报告数据分析困难的问题,提出基于多维字符特征表示设备事故信息抽取方法,在数据预处理阶段,提出主题模式匹配方法,抽取命名实体所属的主题段落;在文本特征表示中,提出多维特征表示方法将文本转化为特征向量;采用长短时记忆网络(BiLSTM)与条件随机场(CRF)神经网络实现铁路设备事故命名实体识别模型训练;采用铁路设备事故调查报告进行试验验证。结果表明:通过主题模式匹配预处理,多维字符特征+BiLSTM+CRF模型的综合评价指标提升22.86%,多维字符特征表示方法相比word2vec特征表示方法,能够使BiLSTM+CRF模型的综合评价指标提升4.89%。

关键词: 多维字符特征, 铁路设备事故, 信息抽取, 主题模式匹配, 命名实体识别

Abstract:

In order to address difficulty in data analysis in investigation reports of railway equipment accidents, an accident information extraction method based on multi-dimensional character feature representation was proposed. Firstly, a subject pattern matching method was put forward for data preprocessing stage to extract subject paragraphs to which named entity belonged. For text feature representation, a multi-dimensional feature representation method was proposed to transform text into feature vector, and training of named entity recognition model was carried out by using bidirection long short term memory(BiLSTM)+ conditional random fields(CRF) neural network. Finally, accident investigation report was used for experimental verification. The results show that the comprehensive evaluation index of multi-dimensional character +BiLSTM+CRF model is improved by 22.86% through preprocessing of subject pattern matching. And compared with word2vec feature representation, multi-dimensional one can improve evaluation index of BiLSTM+CRF model by 4.89%.

Key words: multi-dimensional character feature, railway equipment accident, information extraction, subject pattern matching, named entity recognition.

张鹏翔. 多维字符特征表示的铁路设备事故信息抽取方法[J]. 中国安全科学学报, 2022, 32(6): 109-114.

ZHANG Pengxiang. Information extraction method for railway equipment accidents based on multi-dimensional character feature representation[J]. China Safety Science Journal, 2022, 32(6): 109-114.

图/表 7

表1

图1

图2

图3

图4

表2

表3

参考文献 18

[1]	康高亮. 中国高速铁路安全保障体系研究与实践[J]. 铁道学报, 2017, 39(11):1-7.
	KANG Gaoliang. Research and practice of safety guarantee system for China high-speed railway[J]. Journal of the China Railway Society, 2017, 39(11):1-7.
[2]	许未, 何世伟, 刘朝辉, 等. 基于关联规则的铁路事故致因网络构建与分析[J]. 铁道运输与经济, 2020, 42(11):72-79.
	XU Wei, HE Shiwei, LIU Zhaohui, et al. Construction and analysis of railway accident causation network based on association rules[J]. Railway Transport and Economy, 2020, 42(11):72-79.
[3]	罗文慧, 蔡凤田, 吴初娜, 等. 基于文本挖掘的道路运输安全风险源辨识模型[J]. 西南交通大学学报, 2021, 56(1): 147-152.
	LUO Wenhui, CAI Fengtian, WU Chu'na, et al. Road transportation safety risk source identification model based on text mining[J]. Journal of Southwest Jiaotong University, 2021, 56(1): 147-152.
[4]	王红, 祝寒, 林海舟. 航空安全事故因果关系抽取方法的研究[J]. 计算机工程与应用, 2020, 56(11):265-270.
	WANG Hong, ZHU Han, LIN Haizhou. Research on causality extraction of civil aviation accident[J]. Computer Engineering and Applications, 2020, 56(11): 265-270.
[5]	刘倩文. 海事事故调查报告中介词及介词短语汉译实践报告[D]. 大连: 大连海事大学, 2019.
	LIU Qianwen. Practice report on Chinese translation of prepositions and preposition phrases in marine accident investigation report[D]. Dalian: Dalian Maritime University, 2019.
[6]	周俊生, 戴新宇, 尹存燕, 等. 自然语言信息抽取中的机器学习方法研究[J]. 计算机科学, 2005, 32(3):186-189.
	ZHOU Junsheng, DAI Xinyu, YI Cunyan et al. The methods of machine learning for natural language information extraction[J]. Computer Science, 2005, 32(3):186-189.
[7]	陈玉华. 信息抽取在自然语言查询接口中的应用与研究[J]. 计算机与数字工程, 2008, 36(4):161-163.
	CHEN Yuhua. Application and research on information extraction in natural language query interface[J]. Computer & Digital Engineering, 2008, 36(4):161-163.
[8]	WU Lei, HOI S C H, YU Nenghai. Semantics-preserving bag-of-words models and applications[J]. IEEE Transactions on Image Processing, 2010, 19(7):1908-1920. doi: 10.1109/TIP.2010.2045169 pmid: 20227977
[9]	ZHAI Chengxiang, LAFFERTY J. A study of smoothing methods for language models applied to information retrieval[J]. Acm Transactions on Information Systems, 2004, 22(2):179-214. doi: 10.1145/984321.984322
[10]	TURNEY P D, PANTEL P. From frequency to meaning: vector space models of semantics[J]. Journal of Artificial Intelligence Research, 2010, 37(1):141-188. doi: 10.1613/jair.2934
[11]	杨连报. 铁路事故故障文本大数据分析关键技术研究及应用[D]. 北京: 中国铁道科学研究院, 2018.
	YANG Lianbao. Research of railway fault accident text big data mining key technologies and application[D]. Beijing: China Academy of Railway Sciences, 2018.
[12]	LI Xinqin, SHI Tianyun, LI Ping, et al. BiLSTM-CRF model for named entity recognition in railway accident and fault analysis report[C]. APCIM 2018 and 7^th International Conference on Transportation and Traffic Engineering, 2018:1-5.
[13]	QIN Yi, ZENG Yifei. Research of clinical named entity recognition based on Bi-LSTM-CRF[J]. Journal of Shanghai Jiaotong University:Science, 2018, 23(3):58-63.
[14]	QIN Ya, SHEN Guowei, ZHAO Wenbo, et al. A network security entity recognition method based on feature template and CNN-BiLSTM-CRF[J]. Frontiers of Information Technology & Electronic Engineering, 2019, 20(6):872-884.
[15]	WANG Chenglong, JIANG Feijun, YANG Hongxia. A hybrid framework for text modeling with convolutional RNN[C]. Proceedings of the 23^rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2017:2061-2069.
[16]	LEI Tao, BARZILAY R, JAAKKOLA T S. Molding CNNs for text: non-linear, non-consecutive convolutions[J]. Indiana University Mathematics Journal, 2015, 58(3):1151-1186. doi: 10.1512/iumj.2009.58.3578
[17]	宋敦江, 杨霖, 钟少波. 基于BERT的灾害三元组信息抽取优化研究[J]. 中国安全科学学报, 2022, 32(2):115-120. doi: 10.16265/j.cnki.issn1003-3033.2022.02.016
	SONG Dunjiang, YANG Lin, ZHONG Shaobo. Research on optimization of disaster triplet information extraction based on BERT[J]. China Safety Science Journal, 2022, 32(2):115-120. doi: 10.16265/j.cnki.issn1003-3033.2022.02.016
[18]	冯文刚. 基于深度长短记忆模型的民航安保事件分析[J]. 中国安全科学学报, 2021, 31(9):1-7. doi: 10.16265/j.cnki.issn1003-3033.2021.09.001
	FENG Wengang. Research on civil aviation security event analysis based on deep LSTM model[J]. China Safety Science Journal, 2021, 31(9):1-7. doi: 10.16265/j.cnki.issn1003-3033.2021.09.001

抽取信息	所在主题段落	命名实体类型
发生时间	事故概况	FSSJ
发生辖区	事故概况	FSXQ
车型车号	事故概况	CXCH
故障等级	事故概况	SGDJ
发生原因	原因分析	YYFX
故障现象	原因分析	GZXX
责任单位	事故性质及责任认定	ZRDW

方法	P	R	F₁	耗时/s
主题模式匹配+ 多维字符特征+ BiLSTM+CRF	0.952 3	0.943 5	0.947 9	363
多维字符特征+ BiLSTM+CRF	0.716 2	0.722 5	0.719 3	126

方法	P	R	F₁
主题模式匹配+多维字符特征+BiLSTM+CRF	0.952 3	0.943 5	0.947 9
主题模式匹配+word2vec+ BiLSTM+CRF	0.891 8	0.904 5	0.898 1

多维字符特征表示的铁路设备事故信息抽取方法

Information extraction method for railway equipment accidents based on multi-dimensional character feature representation

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 7

参考文献 18

相关文章 0

编辑推荐

Metrics

本文评价