Abstract：In order to solve data deficiency and excessive staff workload in the risk-source identification of road transportation safety, an automatic identification model is proposed from the angle of text mining. Firstly， the model performs feature enhancement preprocessing operation through the causality sentence extraction and extracted sentence segmentation. Secondly, the feature construction adapted to the convolutional neural network (CNN) is conducted, which contains word information and position information. Thirdly, the results of feature construction feed into the CNN to realize the identification of risk sources. Finally，experiments are conducted with the data sets of traffic accidents, demonstrating that the proposed model can identify most of risk sources for road transportation safety with the accuracy of about 77.321%.
丁辉. 风险、风险源、危险源的关系探讨[J]. 安全,2019,40(4): 1-4DING Hui. Discussion on the relationship among risk,risk source and hazard[J]. Safety & Security, 2019, 40(4): 1-4
[2]
LI J, WANG J, XU N, et al. Importance degree research of safety risk management processes of urban rail transit based on text mining method[J]. Information, 2018, 9(2): 9-26
[3]
SHI D, GUAN J, ZURADA J, et al. A data-mining approach to identification of risk factors in safety management systems[J]. Journal of Management Information Systems, 2017, 34(4): 1054-1081
[4]
WILLIAMS T, BETAK J. A comparison of LSA and LDA for the analysis of railroad accident text[J]. Journal of Ubiquitous System & Pervasive Networks, 2019, 11(1): 11-15
[5]
ANDRZEJCZAK C, KARWOWSKI W, MIKUSINSKI P. Application of diffusion maps to identify human factors of self-reported anomalies in aviation[J]. Work, 2012, 41(S1): 188-197
[6]
TANGUY L, TULECHKI N, URIELI A, et al. Natural language processing for aviation safety reports:from classification to interactive analysis[J]. Computers in Industry, 2016, 78: 80-95
[7]
ZHANG X, MAHADEVAN S. Ensemble machine learning models for aviation incident risk prediction[J]. Decision Support Systems, 2019, 116: 48-63
[8]
AN N, XIAO Y B, YUAN J, et al. Extracting causal relations from the literature with word vector mapping[J]. Computers in Biology and Medicine, 2019, 115: 103524.1-103524.8
[9]
ZENG D，LIU K，LAI S，et al. Relation classification via convolutional deep neural network[C]//Proceedings of the 25th International Conference on Computational Linguistics. New York：Curran Associates，2014：2335-2344.
[10]
袁飞,赵绪言,王一戈,等. 基于轻量级卷积神经网络的烟雾识别算法[J]. 西南交通大学学报,2020,55(5): 1111-1116,1132YUAN Fei, ZHAo Xuyan, WANG Yige, et al. Smoke recognition algorithm based on lightweight convolutional neural network[J]. Journal of Southwest Jiaotong University, 2020, 55(5): 1111-1116,1132
[11]
裘江南. 汉语文本中突发事件因果关系抽取方法研究[D]. 大连：大连理工大学，2012.
[12]
HE B, GUAN Y, DAI R. Classifying medical relations in clinical text via convolutional neural networks[J]. Artificial Intelligence in Medicine, 2019, 93: 43-49
