The paper integrates Rough Sets (RS) and Bayesian Networks (BN) for roadway traffic accident analysis. RS reduction of attributes is first employed to generate the key set of attributes affecting accident outcomes, which are then fed into a BN structure as nodes for BN construction and accident outcome classification. Such RS-based BN framework combines the advantages of RS in knowledge reduction capability and BN in describing interrelationships among different attributes. The framework is demonstrated using the 100-car naturalistic driving data from Virginia Tech Transportation Institute to predict accident type. Comparative evaluation with the baseline BNs shows the RS-based BNs generally have a higher prediction accuracy and lower network complexity while with comparable prediction coverage and receiver operating characteristic curve area, proving that the proposed RS-based BN overall outperforms the BNs with/without traditional feature selection approaches. The proposed RS-based BN indicates the most significant attributes that affect accident types include pre-crash manoeuvre, driver’s attention from forward roadway to centre mirror, number of secondary tasks undertaken, traffic density, and relation to junction, most of which feature pre-crash driver states and driver behaviours that have not been extensively researched in literature, and could give further insight into the nature of traffic accidents.
World Health Organization. Road Traffic Injuries. World Health Organization; 2016. Available from: http://www.who.int/mediacentre/factsheets/fs358/en/
Klauer S, et al. The Impact of Driver Inattention on Near-Crash/Crash Risk: An Analysis Using the 100-Car Naturalistic Driving Study Data. Washington D.C.: U.S. Department of Transportation; 2006.
Piccinini GB, Engström J, Bärgman J, et al. Factors contributing to commercial vehicle rear-end conflicts in China: A study using on-board event data recorders. Journal of Safety Research. 2017;62: 143-153.
Wang W, Liu C, Zhao D. How Much Data is Enough? A Statistical Approach with Case Study on Longitudinal Driving Behavior. IEEE Transactions on Intelligent Vehicles. 2017;2(2): 85-98.
Precht L, Keinath A, Krems JF. Identifying the main factors contributing to driving errors and traffic violations – Results from naturalistic driving data. Transportation Research Part F: Traffic Psychology & Behaviour. 2017;49: 49-92.
Xiong H, Bao S, Sayer J, et al. Examination of drivers' cell phone use behavior at intersections by using naturalistic driving data. Journal of Safety Research. 2015;54: 89.e29-93.
Precht L, Keinath A, Krems J F. Effects of driving anger on driver behavior – Results from naturalistic driving data. Transportation Research Part F: Psychology & Behaviour. 2017;45: 75-92.
Wu KF, Aguerovalverde J, Jovanis PP. Using naturalistic driving data to explore the association between traffic safety-related events and crash risk at driver level. Accident Analysis & Prevention. 2014;72: 210-218.
Wang Y, Zhang W. Analysis of Roadway and Environmental Factors Affecting Traffic Crash Severities. Transportation Research Procedia. 2017;25: 2124-2130.
Zhang G, Yau KK, Zhang X, et al. Traffic accidents involving fatigue driving and their extent of casualties. Accident Analysis & Prevention. 2016;87: 34-42.
Talbot R, Fagerlind H, Morris A. Exploring inattention and distraction in the SafetyNet Accident Causation Database. Accident Analysis & Prevention. 2013;60(45): 445-455.
Chang L, Wang H. Analysis of traffic injury severity: An application of non-parametric classification tree techniques. Accident Analysis & Prevention. 2006;38(5): 1019-1027.
Zong F, Xu H, Zhang H. Prediction for Traffic Accident Severity: Comparing the Bayesian Network and Regression Models. Mathematical Problems in Engineering. 2013;2013(2-3): 206-226.
Pawlak Z. Rough sets. International Journal of Parallel Programming. 1982;11(5): 341-356.
Bello R, Falcon R. Rough Sets in Machine Learning: A Review. Thriving Rough Sets, Studies in Computational Intelligence. 2017;708: 87-118.
Wong J, Chung Y. Rough set approach for accident chains exploration. Accident Analysis & Prevention. 2007;39(3): 629-637.
Peng L, Chao-Zhong W, Huang Z. Situation Assessment of Vehicle Collision Risk Based on Variable Precision Rough Set. Journal of Transportation Systems Engineering & Information Technology. 2013;13(5): 120-126.
Simoncic M. A Bayesian network model of two-car accidents. Journal of Transportation & Statistics. 2004;7(1): 3594-3597.
Karimnezhad A, Moradi F. Road accident data analysis using Bayesian networks. Transportation Letters: The International Journal of Transportation Research. 2015.
Oña J, López G, Mujalli R, et al. Analysis of traffic accidents on rural highways using Latent Class Clustering and Bayesian Networks. Accident Analysis & Prevention. 2013;51C(2): 1-10.
Oña J, Mujalli R, Calvo F. Analysis of traffic accident injury severity on Spanish rural highways using Bayesian networks. Accident Analysis & Prevention. 2011;43(1): 402-11.
Mujalli R, De O. A method for simplifying the analysis of traffic accidents injury severity on two-lane highways using Bayesian networks. Journal of Safety Research. 2011;42(5): 317-326.
Guo X, Zhang H, Fang Z. Bayesian network modeling for causation analysis of traffic accident. Journal of Jilin University (Engineering and Technology Edition). 2011;41: 89-94.
Bishop C. Pattern Recognition and Machine Learning. New York: Springer-Verlag; 2006.
Cios K, Pedrycz W, Świniarski R. Data Mining Methods in Knowledge Discovery. London: Kluwer Academic Publishers; 1998.
John G, Kohavi R, Pfleger K. Irrelevant features and the subset selection problem. In: Cohen WW, Hirsh H. (eds.) Machine Learning: Proceedings of the 11th International Conference of Machine Learning, July 10–13 1994, Rutgers University, New Brunswick, NJ. San Francisco CA: Morgan Kaufmann Publishers; 1994. p. 121-129.
Blanco R, Inza I, Merino M, et al. Feature selection in Bayesian classifiers for the prognosis of survival of cirrhotic patients treated with TIPS. Journal of Biomedical Informatics. 2005;38(5): 376-388.
Neapolitan R. Probabilistic Methods for Bioinformatics. San Francisco: Morgan Kaufmann Publishers; 2009.
Madden M. On the classification performance of TAN and general Bayesian networks. Knowledge-Based Systems. 2009;22(7): 489-495.
Cheng L, Chen X, Wei M, et al. Modeling mode choice behavior incorporating household and individual sociodemographics and travel attributes based on rough sets theory. Computational Intelligence & Neuroscience. 2014;2014: 1-9.
Virginia Tech Transportation Institute. VTTI Data Warehouse. Virginia Tech Transportation Institute; 2016. Available from: http://forums.vtti.vt.edu/index.php?/files/category/3-100-car-data/
Guo F. Near-Crashes as Crash Surrogate for Naturalistic Driving Studies. Transportation Research Record Journal of the Transportation Research Board. 2010;2147(-1): 66-74.
Dingus T, Hulse M, Antin J, et al. Attentional demand requirements of an automobile moving-map navigation system. Transportation Research Part A: Policy and Practice. 1989;23(4): 301-315.
Komorowski E. ROSETTA- A Rough Set Toolkit for Analysis of Data. Proceedings of 3rd International Joint Conference on Information Sciences; 1997. p. 303-407.
Witten I, Frank E. Data Mining: Practical Machine Learning Tools and Techniques. 2nd ed. San Francisco: Morgan Kaufmann; 2005.
Guest Editor: Eleonora Papadimitriou, PhD
Editors: Dario Babić, PhD; Marko Matulin, PhD; Marko Ševrović, PhD.
The mission of our journal is to provide high quality communication of research results, which is crucial for the effectiveness of scientific research, the promotion of scientific information, new ideas, and original research in the field of traffic and transportation engineering.
1848-4069
0353-5320
Proud publisher
Accelerating Discoveries in Traffic Science |
2024 © Promet - Traffic&Transportation journal
