基于选择集成的山区高速事故预测模型

doi:10.13229/j.cnki.jdxbgxb.20230725

Abstract

Abstract:

To improve the prediction accuracy and reduce the robustness of the traffic accident prediction model， this paper uses the Stacking integration strategy to construct an integrated traffic accident prediction model. Firstly， single traffic accident prediction models based on eight machine learning models， such as Decision Tree and Extra Tree， were constructed and the MIC test was used to measure the similarity of each traffic prediction model with the graph coloring method， and the models with low similarity and high diversity were selected to participate in the integration. Secondly， Box-Cox transformations were applied to the results of the single accident prediction models and different weights were assigned to each single model separately using feature weighting method. Finally， models such as BP neural network and Logistic regression were selected as meta-learners for Stacking integration. The results of the study show that the prediction accuracy of the integrated model with BP neural network selected for the meta-learner is higher than other integrated models， and the MAE and RMSE of the integrated model have been respectively reduced by 24% and 14% and the R² has been improved by 6% compared to the single accident prediction model with the highest prediction accuracy.

Key words: transportation planning and management, traffic accident prediction, mountain highways, machine learning, integrated learning

CLC Number:

U491.31

Xiang-hai MENG,Guo-rui WANG,Ming-yang ZHANG,Bi-jiang TIAN. Traffic accident prediction model of mountain highways based on selection integration[J].Journal of Jilin University(Engineering and Technology Edition), 2025, 55(4): 1298-1306.

Figures/Tables 14

Table 1

Table 2

Table 3

Table 4

Fig.1

Table 5

Table 6

Fig.2

Table 7

Fig.3

Fig.4

Fig.5

Fig.6

Table 8

References 17

1	张显强, 贺中华, 梁永娜, 等. 贵州省道路分形特征及其对交通事故影响机制[J]. 公路, 2017, 62(6):197-203.
	Zhang Xian-qiang, He Zhong-hua, Liang Yong-na, et al. Fractal characteristics of road and its impact mechanism on traffic accidents in Guizhou Province[J]. Highway, 2017, 62(6): 197-203.
2	Macedo M R, Maia M L A, Rabbani E R K, et al. Traffic accident prediction model for rural highways in Pernambuco[J]. Case Studies on Transport Policy, 2022, 10(1): 278-286.
3	马壮林, 邵春福, 李霞. 基于Logistic模型的公路隧道交通事故严重程度的影响因素[J]. 吉林大学学报:工学版,2010, 40(2): 423-426.
	Ma Zhuang-lin, Shao Chun-fu, Li Xia. Analysis of factors affecting accident severity in highway tunnels based on Logistic model[J]. Journal of Jilin University(Engineering and Technology Edition), 2010, 40(2): 423-426.
4	陈英, 袁华智, 黄中祥, 等. 零截尾负二项模型在交叉口事故预测中的应用[J]. 中国公路学报, 2020, 33(4): 146-154.
	Chen Ying, Yuan Hua-zhi, Huang Zhong-xiang, et al. Modeling intersection traffic crashes using a zero-truncated negative binomial model[J]. China Journal of Highway and Transport, 2020, 33(4): 146-154.
5	Roland J, Way P D, Firat C, et al. Modeling and predicting vehicle accident occurrence in Chattanooga, Tennessee[J]. Accident Analysis & Prevention, 2021(149): 105-117.
6	Ihueze C C, Onwurah U O. Road traffic accidents prediction modelling: an analysis of Anambra State, Nigeria[J]. Accident Analysis & Prevention, 2018(7), 112: 21-29.
7	谢学斌, 孔令燕. 基于ARIMA和XGBoost组合模型的交通事故预测[J]. 安全与环境学报, 2021, 21(1):277-284.
	Xie Xue-bin, Kong Ling-yan. On the ways to the traffic accident prediction based on the ARIMA and XGBoost combined model[J]. Journal of Safety and Environment, 2021, 21(1): 277-284.
8	纪俊红, 昌润琪, 温廷新. 基于GSK-AdaBoost-LightGBM的交通事故死亡人数预测研究[J]. 安全与环境工程, 2021, 28(1): 24-28.
	Ji Jun-hong, Chang Run-qi, Wen Ting-xin. Prediction of traffic accident death toll based on GSK-AdaBoost-LightGBM[J]. Safety and Environmental Engineering, 2021, 28(1): 24-28.
9	Vilaa M, Macedo E, Coelho M C. A rare event modelling approach to assess injury severity risk of vulnerable road users[J]. Safety, 2019, 5(2): 29-38.
10	Xing L, He J, Li Y, et al. Comparison of different models for evaluating vehicle collision risks at upstream diverging area of toll plaza[J]. Accident Analysis and Prevention, 2020(135): 86-97.
11	Kwon O H, Rhee W, Yoon Y, et al. Application of classification algorithms for analysis of road safety risk factor dependencies[J]. Accident Analysis and Prevention, 2015(75): 1-15.
12	Zeng K H, Chou S H, Chan F H, et al. Agent-centric risk assessment: accident anticipation and risky region localization[C]∥Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, USA, 2017: 2222-2230.
13	宁静, 佘红艳, 赵东, 等. 一种路网级交通事故风险预测方法[J]. 北京邮电大学学报, 2022, 45(2): 72-78.
	Ning Jing, She Hong-yan, Zhao Dong, et al. A road-level traffic accident risk prediction method[J]. Journal of Beijing University of Posts and Telecommunications, 2022, 45(2): 72-78.
14	Lin L, Wang Q, Sadek A W. A novel variable selection method based on frequent pattern tree for real-time traffic accident risk prediction[J]. Transportation Research Part C: Emerging Technologies, 2015(55): 444-459.
15	孙棣华, 唐亮, 付青松, 等. 基于量子神经网络的道路交通事故预测[J]. 交通运输系统工程与信息, 2010, 10(5): 104-109.
	Sun Di-hua, Tang Liang, Fu Qing-song, et al. Road traffic accidents forecasting based on quantum neural network[J]. Journal of Transportation Systems Engineering and Information Technology, 2010, 10(5): 104-109.
16	覃薇. 基于负二项回归分析的高速公路神经网络事故预测模型[D]. 哈尔滨: 哈尔滨工业大学交通科学与工程学院, 2017.
	Qin Wei. Neural network crash prediction model of freeway based on negative binomial regression analysis[D]. Harbin: School of Transportation Science and Engineering of Harbin Institute of Technology, 2017.
17	范中洲, 赵羿, 周宁, 等. 基于灰色BP神经网络组合模型的水上交通事故数预测[J]. 安全与环境学报, 2020, 20(3): 857-861.
	Fan Zhong-zhou, Zhao Yi, Zhou Ning, et al. Integrated model for forecasting waterway traffic accidents based on the Gray-BP neural network[J]. Journal of Safety and Environment, 2020, 20(3): 857-861.

Related Articles 15

[1]	Qing-liang JIN,Xin-sen ZHOU,Yi CHEN,Cheng-wen WU. Predictive model for identifying innovative university talents based on the swarm intelligence evolution enhanced kernel extreme learning machine [J]. Journal of Jilin University(Engineering and Technology Edition), 2025, 55(5): 1763-1771.
[2]	Xiu-zhi ZHAO,De-hong XIE. Discrimination method for Pu-er tea varieties based on noise-robust feature extraction [J]. Journal of Jilin University(Engineering and Technology Edition), 2025, 55(5): 1756-1762.
[3]	Sheng-qi MEI,Xiao-dong LIU,Xing-ju WANG,Xu-feng LI,Teng WU,Xiang-xu CHENG. Prediction of high strength concrete creep based on parametric MIC analysis and machine learning algorithm [J]. Journal of Jilin University(Engineering and Technology Edition), 2025, 55(5): 1595-1603.
[4]	Hao YUE,Xiao CHANG,Jian-ye LIU,Qiu-shi QU. Customized bus route optimization with vehicle window [J]. Journal of Jilin University(Engineering and Technology Edition), 2025, 55(4): 1266-1274.
[5]	Wen-jing WU,Chun-chun DENG,Hong-fei JIA,Shu-hang SUN. Evaluation of road network unblocked reliability and identification of critical sections under influence of flooding [J]. Journal of Jilin University(Engineering and Technology Edition), 2025, 55(4): 1250-1257.
[6]	Yin-fei DAI,Xiu-zhen ZHOU,Yu-bao LIU,Zhi-yuan LIU. In⁃vehicle network intrusion detection system based on CAN bus data [J]. Journal of Jilin University(Engineering and Technology Edition), 2025, 55(3): 857-865.
[7]	Shu-hong MA,Jun-jie ZHANG,Xi-fang CHEN,Guo-mei LIAO. Identifying urban functional structures using time-series taxi data [J]. Journal of Jilin University(Engineering and Technology Edition), 2025, 55(2): 603-613.
[8]	Tian-yang GAO,Da-wei HU,Rui-sen JIANG,Xue WU,Hui-tian LIU. Optimization study of zonal-based flexible feeder bus routes based on modular vehicle system [J]. Journal of Jilin University(Engineering and Technology Edition), 2025, 55(2): 537-545.
[9]	Yu-ran LI,Fei WANG,Cai-hua ZHU,Fei HAN,Yan LI. Chain-effect utility of factors influencing residents' commuting mode choice in polluted weather [J]. Journal of Jilin University(Engineering and Technology Edition), 2025, 55(2): 577-590.
[10]	Hui-zhi XU,Shi-sen JIANG,Xiu-qing WANG,Shuang CHEN. Vehicle target detection and ranging in vehicle image based on deep learning [J]. Journal of Jilin University(Engineering and Technology Edition), 2025, 55(1): 185-197.
[11]	Lei ZHANG,Jing JIAO,Bo-xin LI,Yan-jie ZHOU. Large capacity semi structured data extraction algorithm combining machine learning and deep learning [J]. Journal of Jilin University(Engineering and Technology Edition), 2024, 54(9): 2631-2637.
[12]	Chang-jiang ZHENG,Tong-tong TAO,Zhi-chao CHEN. Cascading failure model based on adjustable redistribution of traffic flow [J]. Journal of Jilin University(Engineering and Technology Edition), 2024, 54(9): 2441-2450.
[13]	Xiao-yue WEN,Guo-min QIAN,Hua-hua KONG,Yue-jie MIU,Dian-hai WANG. TrafficPro： a framework to predict link speeds on signalized urban traffic network [J]. Journal of Jilin University(Engineering and Technology Edition), 2024, 54(8): 2214-2222.
[14]	Yun-juan YAN,Wei-xiong ZHA,Jun-gang SHI,Li-ping YAN. Double layer optimization model of charging pile based on random charging demand [J]. Journal of Jilin University(Engineering and Technology Edition), 2024, 54(8): 2238-2244.
[15]	Da-yi QU,Hao-min LIU,Zi-yi YANG,Shou-chen DAI. Dynamic allocation mechanism and model of traffic flow in bottleneck section based on vehicle infrastructure cooperation [J]. Journal of Jilin University(Engineering and Technology Edition), 2024, 54(8): 2187-2196.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

Comments

Recommended 10

[1]	LI Shoutao, LI Yuanchun. Autonomous Mobile Robot Control Algorithm Based on Hierarchical Fuzzy Behaviors in Unknown Environments[J]. 吉林大学学报(工学版), 2005, 35(04): 391 -397 .
[2]	Liu Qing-min，Wang Long-shan，Chen Xiang-wei，Li Guo-fa. Ball nut detection by machine vision[J]. 吉林大学学报(工学版), 2006, 36(04): 534 -538 .
[3]	Li Hong-ying; Shi Wei-guang;Gan Shu-cai. Electromagnetic properties and microwave absorbing property of Z type hexaferrite Ba_3-xLa_xCo₂Fe₂₄O₄₁[J]. 吉林大学学报(工学版), 2006, 36(06): 856 -0860 .
[4]	Yang Shu-kai, Song Chuan-xue, An Xiao-juan, Cai Zhang-lin . Analyzing effects of suspension bushing elasticity on vehicle yaw response character with virtual prototype method[J]. 吉林大学学报(工学版), 2007, 37(05): 994 -0999 .
[5]	. [J]. 吉林大学学报(工学版), 2007, 37(06): 1284 -1287 .
[6]	Che Xiang-jiu，Liu Da-you，Wang Zheng-xuan . Construction of joining surface with G¹ continuity for two NURBS surfaces[J]. 吉林大学学报(工学版), 2007, 37(04): 838 -841 .
[7]	Liu Han-bing, Jiao Yu-ling, Liang Chun-yu,Qin Wei-jun . Effect of shape function on computing precision in meshless methods[J]. 吉林大学学报(工学版), 2007, 37(03): 715 -0720 .
[8]	Zhang Quan-fa，Li Ming-zhe，Sun Gang，Ge Xin . Comparison between flexible and rigid blank-holding in multi-point forming[J]. 吉林大学学报(工学版), 2007, 37(01): 25 -30 .
[9]	. [J]. 吉林大学学报(工学版), 2007, 37(04): 0 .
[10]	Li Yue-ying，Liu Yong-bing，Chen Hua . Surface hardening and tribological properties of a cam materials[J]. 吉林大学学报(工学版), 2007, 37(05): 1064 -1068 .

变量标识	变量名称	单位	最大值	最小值	平均值	标准差
NA	年平均事故次数	起	15.50	0	1.22	1.89
DT	日交通量	pcu/day	14 797	6 390	1 0246	2792
L	路段长度	m	598.36	50.18	244.00	135.97
LL	直线段长度	m	3 796.49	0	395.82	772.36
AH	平曲线偏角	（°）	185.68	0	23.60	29.05
CH	平曲线曲率	1/km	7.69	0	1.20	1.50
LH	平曲线长度	m	2 299.86	0	326.87	383.78
LS	纵坡长度	m	4 199.84	10	616.97	616.54
SD	竖曲线坡度差	（°）	8.15	0	1.30	1.70
ASC	当前累积坡度	%	0	-4.98	-0.64	1.23
LSC	当前累积坡长	m	26 917.28	0	2 029.34	4 881.48
AS	纵坡坡度	（°）	6.00	-6.00	0	2.25

变量标识/赋值	变量名称	百分比 /%	变量表示	变量名称	百分比 /%
HC	平曲线		VT	竖曲线类型
0	直线路段	41.34	0	纵坡路段	61.41
-1	右偏	29.33	-1	凸型竖曲线	17.93
1	左偏	29.33	1	凹型竖曲线	20.66
TC	缓和曲线		RT	路段类型
0	直线路段	41.34	1	基本路段	83.21
1	有	53.35	2	收费站	10.33
-1	无	5.31	3	服务区	5.31
CS	连续下坡		4	桥梁	1.15
1	有	25.91
0	无	74.09

单一模型	MAE	RMSE	R²
决策树	0.34	0.59	0.74
ET	0.32	0.52	0.8
RF	0.35	0.61	0.74
GBDT	0.35	0.51	0.81
KNN	0.43	0.69	0.64
XGBoost	0.29	0.43	0.84
LightGBM	0.33	0.51	0.81
CatBoost	0.3	0.46	0.83

单一事故预测模型	决策树模型	ET 模型	RF 模型	GBDT 模型	KNN 模型	XGBoost 模型	LightGBM 模型	CatBoost 模型
决策树模型	1	0.97	0.88	0.94	0.95	0.96	0.82	0.69
ET模型	0.97	1	0.93	0.99	0.97	0.99	0.87	0.73
RF模型	0.88	0.93	1	0.92	0.9	0.95	0.87	0.82
GBDT模型	0.94	0.99	0.92	1	0.96	0.98	0.87	0.76
KNN模型	0.95	0.97	0.9	0.96	1	0.96	0.82	0.71
XGBoost模型	0.96	0.99	0.95	0.98	0.96	1	0.9	0.77
LightGBM模型	0.82	0.87	0.87	0.87	0.82	0.9	1	0.76
CatBoost模型	0.69	0.73	0.82	0.76	0.71	0.77	0.76	1

单一事故预测模型	决策树模型	ET 模型	RF 模型	GBDT 模型	KNN 模型	XGBoost 模型	LightGBM 模型	CatBoost 模型
决策树模型	0	0	0	0	0	0	1	1
ET模型	0	0	0	0	0	0	1	1
RF模型	0	0	0	0	0	0	1	1
GBDT模型	0	0	0	0	0	0	1	1
KNN 模型	0	0	0	0	0	0	1	1
XGBoost模型	0	0	0	0	0	0	0	1
LightGBM模型	1	1	1	1	1	0	0	1
CatBoost模型	1	1	1	1	1	1	1	0

Traffic accident prediction model of mountain highways based on selection integration

RICH HTML

PDF (PC)