Predicting and explaining severity of road accident using artificial intelligence techniques, SHAP and feature analysis

References

• WHO. Global status report on road safety 2018 summary. Geneva: World Health Organization; 2018. (WHO/NMH/NVI/18.20) License: CC BY-NC-SA 3.0 IGO).
   Google Scholar
• Sustainable Development Goals. Road safety for all; 2019 Jan. [cited 2021 Jun 6]. Available from: https://unece.org/DAM/trans/roadsafe/publications/Road_Safety_for_All.pdf
   Google Scholar
• WHO. Road traffic injuries. World Health Organization; 2020. [cited 2021 Jun 6]. Available from: https://www.who.int/news-room/fact-sheets/detail/road-traffic-injuries
   Google Scholar
• Chen S, Kuhn M, Prettner K, et al. The global macroeconomic burden of road injuries: estimates and projections for 166 countries. Lancet Planet Health. 2019;3(9):e390–e398.
   PubMedGoogle Scholar
• Birek L, Grzywaczewski A, Iqbal R, et al. A novel big data analytics and intelligent technique to predict driver's intent. Comput Ind. 2018;99:226–240.
   Web of Science ®Google Scholar
• Rovsek V, Batista M, Bogunović B. Identifying the key risk factors of traffic accident injury severity on Slovenian roads using a non-parametric classification tree. Transport. 2014;32(3):272–281.
   Google Scholar
• Das A, Khan MN, Ahmed MM. Detecting lane change maneuvers using SHRP2 naturalistic driving data: a comparative study machine learning techniques. Accid Anal Prev. 2020;142:105578.
   PubMed Web of Science ®Google Scholar
• Yahaya M, Fan W, Fu C, et al. A machine-learning method for improving crash injury severity analysis: a case study of work zone crashes in Cairo. Int J Inj Contr Saf Promot. 2020;27(3):266–275.
   PubMed Web of Science ®Google Scholar
• Ba Y, Zhang W, Wang Q, et al. Crash prediction with behavioral and physiological features for advanced vehicle collision avoidance system. Transp Res C: Emerg Technol. 2017;74:22–33.
   Web of Science ®Google Scholar
• Allen T, Newstead S, Lenné MG, et al. Contributing factors to motorcycle injury crashes in Victoria, Australia. Transp Res F: Traffic Psychol Behav. 2017;45:157–168.
   Web of Science ®Google Scholar
• Michalaki P, Quddus MA, Pitfield D, et al. Exploring the factors affecting motorway accident severity in England using the generalised ordered logistic regression model. J Safety Res. 2015;55:89–97.
   PubMed Web of Science ®Google Scholar
• Ahmadi E, Süer GA, Al-Ogaili F. Solving stochastic shortest distance path problem by using genetic algorithms. Proc Comput Sci. 2018;140:79–86.
   Google Scholar
• Pillajo-Quijia G, Arenas-Ramírez B, González-Fernández C, et al. Influential factors on injury severity for drivers of light trucks and vans with machine learning methods. Sustainability. 2020;12(4):1324. 2–28.
   Google Scholar
• Panicker AK, Ramadurai G. Injury severity prediction model for two-wheeler crashes at mid-block road sections. Int J Crashworthiness. 2020:1–9.
   Google Scholar
• Fiorentini N, Losa M. Handling imbalanced data in road crash severity prediction by machine learning algorithms. Infrastructures. 2020;5(7):61. 1–24.
   Google Scholar
• Chen MM, Chen MC. Modeling road accident severity with comparisons of logistic regression, decision tree and random Forest. Information. 2020;11(5):270.
   Google Scholar
• Ahmadi A, Jahangiri A, Berardi V, et al. Crash severity analysis of rear-end crashes in California using statistical and machine learning classification methods. J Transp Saf Secur. 2020;12(4):522–546.
   Web of Science ®Google Scholar
• Das S, Datta S, Zubaidi HA, et al. Applying interpretable machine learning to classify tree and utility pole related crash injury types. IATSS Res. 2021;45(3):310–316.
   Google Scholar
• Kumar S, Toshniwal D. Severity analysis of powered two wheeler traffic accidents in Uttarakhand, India. Eur Transp Res Rev. 2017;9(2):1–10.
   Google Scholar
• Lee J, Yeo J, Yun I, et al. Factors affecting crash involvement of commercial vehicle drivers: evaluation of commercial vehicle drivers' characteristics in South Korea. J Adv Transp. 2020;2020:1–8.
   Google Scholar
• Wu W, Jiang S, Liu R, et al. Economic development, demographic characteristics, road network and traffic accidents in Zhongshan, China: gradient boosting decision tree model. Transp A: Transp Sci. 2020;16(3):359–387.
   Web of Science ®Google Scholar
• Parsa AB, Movahedi A, Taghipour H, et al. Toward safer highways, application of XGBoost and SHAP for real-time accident detection and feature analysis. Accid Anal Prev. 2020;136:105405.
   PubMed Web of Science ®Google Scholar
• Ren H, Song Y, Liu J, et al. A deep learning approach to the prediction of short-term traffic accident risk. arXiv preprint arXiv:1710.09543. 2017.
   Google Scholar
• Harb R, Yan X, Radwan E, et al. Exploring precrash maneuvers using classification trees and random forests. Accid Anal Prev. 2009;41(1):98–107.
   PubMed Web of Science ®Google Scholar
• Sangare M, Gupta S, Bouzefrane S, et al. Exploring the forecasting approach for road accidents: analytical measures with hybrid machine learning. Expert Syst Appl. 2021;167:113855.
   Web of Science ®Google Scholar
• Leevy JL, Khoshgoftaar TM, Bauder RA, et al. A survey on addressing high-class imbalance in big data. J Big Data. 2018;5(1):1–30.
   Google Scholar
• Abou Elassad ZE, Mousannif H, Al Moatassime H. A proactive decision support system for predicting traffic crash events: a critical analysis of imbalanced class distribution. Knowl Based Syst. 2020;205:106314.
   Web of Science ®Google Scholar
• Kaur P, Gosain A. Comparing the behavior of oversampling and undersampling approach of class imbalance learning by combining class imbalance problem with noise. In: ICT based innovations. Singapore: Springer; 2018. p. 23–30.
   Google Scholar
• Yu R, Abdel-Aty M. Utilizing support vector machine in real-time crash risk evaluation. Accid Anal Prev. 2013;51:252–259.
   PubMed Web of Science ®Google Scholar
• Adanu EK, Hainen A, Jones S. Latent class analysis of factors that influence weekday and weekend single-vehicle crash severities. Accid Anal Prev. 2018;113:187–192.
   PubMed Web of Science ®Google Scholar
• Zhang Z, He Q, Gao J, et al. A deep learning approach for detecting traffic accidents from social media data. Transp Res C: Emerg Technol. 2018;86:580–596.
   Web of Science ®Google Scholar
• Salam S, Kim D, Ahmed F, et al. Exploring the roles of social media data to identify the locations and severity of road traffic accidents. J Biomed Health Inform. 2017:1–11.
   Google Scholar
• United Nations. (UN). Transforming our World: the 2030 agenda for sustainable development. [cited 2019 Dec 5). Available from: https://sustainabledevelopment.un.org/post2015/summit
   Google Scholar
• Linardatos P, Papastefanopoulos V, Kotsiantis S. Explainable AI: a review of machine learning interpretability methods. Entropy. 2020;23(1):18.
   PubMed Web of Science ®Google Scholar
• World Bank. Together for road safety in South Asia; 2020. [cited 2021 Jun 6]. Available from: https://www.worldbank.org/en/who-we-are/news/campaigns/2020/south-asia-road-safety
   Google Scholar
• Hou R, Kong Y, Cai B, et al. Unstructured big data analysis algorithm and simulation of internet of things based on machine learning. Neural Comput Appl. 2020;32(10):5399–5407.
   Google Scholar
• Minaee S, Kalchbrenner N, Cambria E, et al. Deep learning-based text classification: a comprehensive review. ACM Comput Surv. 2021;54(3):1–40.
   Web of Science ®Google Scholar
• Meng Y, Yang N, Qian Z, et al. What makes an online review more helpful: an interpretation framework using XGBoost and SHAP values. JTAER. 2020;16(3):466–490.
   Google Scholar
• Gramegna A, Giudici P. SHAP and LIME: an evaluation of discriminative power in credit risk. Front Artif Intell. 2021;140.
   Google Scholar
• Lundberg S, Lee S. A unified approach to interpreting model predictions. Adv Neural Inform Process Syst. 2017;30:4765–4774.
   Google Scholar
• Boser BE, Guyon IM, Vapnik VN. A training algorithm for optimal margin classifiers In Proceedings of the Fifth Annual Workshop on Computational Learning Theory; 1992;29:144–152.
   Google Scholar
• Cortes C, Vapnik V. Support-vector networks. Mach Learn. 1995;20(3):273–297.
   Web of Science ®Google Scholar
• Dong N, Huang H, Zheng L. Support vector machine in crash prediction at the level of traffic analysis zones: assessing the spatial proximity effects. Accid Anal Prev. 2015;82:192–198.
   PubMed Web of Science ®Google Scholar
• Wu KP, Wang SD. Choosing the kernel parameters for support vector machines by the inter-cluster distance in the feature space. Pattern Recognit. 2009;42(5):710–717.
   Web of Science ®Google Scholar
• Huang CL, Wang CJ. A GA-based feature selection and parameters optimization for support vector machines. Expert Syst Appl. 2006;31(2):231–240.
   Web of Science ®Google Scholar
• Breiman L. Random forests. Mach Learn. 2001;45(1):5–32.
   Web of Science ®Google Scholar
• Friedman JH. Greedy function approximation: a gradient boosting machine. Ann Stat. 2001:1189–1232.
   Web of Science ®Google Scholar
• Friedman JH. Stochastic gradient boosting. Comput Stat Data Anal. 2002;38(4):367–378.
   Web of Science ®Google Scholar
• Chen T, Guestrin C. Xgboost: a scalable tree boosting system In Proceedings of the 22nd ACM Sigkdd International Conference on Knowledge Discovery and Data Mining; 2016 Aug. p. 785–794.
   Google Scholar
• MORTH report. [cited 2019 May]. Available from: https://morth.nic.in/road-accident-in-india
   Google Scholar
• Dong Y, Peng CYJ. Principled missing data methods for researchers. Springer Plus. 2013;2(1):1–17.
   PubMedGoogle Scholar
• Liu J, Li J, Wang K, et al. Exploring factors affecting the severity of night-time vehicle accidents under low illumination conditions. Adv Mech Eng. 2019;11(4):168781401984094.
   Web of Science ®Google Scholar
• Ramadhan MM, Sitanggang IS, Nasution FR, et al. Parameter tuning in random Forest based on grid search method for gender classification based on voice frequency. DTCSE. 2017.
   Google Scholar
• Delen D, Tomak L, Topuz K, et al. Investigating injury severity risk factors in automobile crashes with predictive analytics and sensitivity analysis methods. J Transp Health. 2017;4:118–131.
   Web of Science ®Google Scholar
• Jianfeng X, Hongyu G, Jian T, et al. A classification and recognition model for the severity of road traffic accident. Adv Mech Eng. 2019;11(5):168781401985189.
   Google Scholar
• Wang Y, Prato CG. Determinants of injury severity for truck crashes on Mountain expressways in China: a case-study with a partial proportional odds model. Saf Sci. 2019;117:100–107.
   Web of Science ®Google Scholar
• Asare IO, Mensah AC. Crash severity modelling using ordinal logistic regression approach. Int J Inj Contr Saf Promot. 2020;27(4):412–419.
   PubMed Web of Science ®Google Scholar
• Bener A, Yildirim E, Özkan T, et al. Driver sleepiness, fatigue, careless behavior and risk of motor vehicle crash and injury: population based case and control study. J Traffic Transp Eng (Engl Ed). 2017;4(5):496–502.
   Google Scholar
• Casado-Sanz N, Guirao B, Gálvez-Pérez D. Population ageing and rural road accidents: analysis of accident severity in traffic crashes with older pedestrians on Spanish crosstown roads. Res Transp Bus Manag. 2019;30:100377.
   Web of Science ®Google Scholar
• Bertoli P, Grembi V. The life‐saving effect of hospital proximity. Health Econ. 2017;26:78–91.
   PubMedGoogle Scholar
• Ghisolfi V, Ribeiro GM, Chaves GDLD, et al. Evaluating impacts of overweight in road freight transportation: a case study in Brazil with system dynamics. Sustainability. 2019;11(11):3128.
   Google Scholar
• Li Y, Fan WD. Modelling severity of pedestrian-injury in pedestrian-vehicle crashes with latent class clustering and partial proportional odds model: a case study of North Carolina. Accid Anal Prev. 2019;131:284–296.
   PubMed Web of Science ®Google Scholar
• Kumar IE, Venkatasubramanian S, Scheidegger C, et al. Problems with shapley-value-based explanations as feature importance measures. In International Conference on Machine Learning. PMLR; 2020 Nov; p. 5491–5500.
   Google Scholar