ML-based Road Accident Hotspot Detection

References

[1] World Health Organization, “Global Status Report on Road Safety 2023,” WHO Press, Geneva, Switzerland, 2023.

[2] R. Elvik, “The predictive validity of empirical Bayes estimates of road safety,” Accident Analysis & Prevention, vol. 40, no. 6, pp. 1964–1969, 2008.

[3] Z. Xie and J. Yan, “Kernel density estimation of traffic accidents in a network space,” Computers, Environment and Urban Systems, vol. 32, no. 5, pp. 396–406, 2008.

[4] C. Chen, G. Zhang, J. Huang, J. Liu, and Y. Qiu, “Examining driver injury severity outcomes in rural non-interstate roadway crashes using a hierarchical ordered logit model,” Accident Analysis & Prevention, vol. 129, pp. 196–204, 2019.

[5] C. Yuan, Y. Li, M. Huang, and S. Wang, “Using traffic flow character-istics to predict real-time conflict risk: A novel method for trajectory data analysis,” Analytic Methods in Accident Research, vol. 35, art. no. 102217, 2022.

[6] A. B. Parsa, A. Taghipour, S. Derrible, and A. K. Mohammadian, “Real-time accident detection: Coping with imbalanced data,” Accident Analysis & Prevention, vol. 129, pp. 202–210, 2019.

[7] S. Jiang, W. Chen, and X. Wang, “Deep learning for traffic accident risk prediction: A convolutional approach on urban grid data,” Transportation Research Part C: Emerging Technologies, vol. 147, art. no. 104004, 2023.

[8] S. M. Lundberg and S.-I. Lee, “A unified approach to interpreting model predictions,” in Proc. 31st Conf. Neural Information Processing Systems (NeurIPS), Long Beach, CA, USA, 2017, pp. 4765–4774.

[9] L. Breiman, “Random forests,” Machine Learning, vol. 45, no. 1, pp. 5–32, 2001.

[10] T. Chen and C. Guestrin, “XGBoost: A scalable tree boosting system,” in Proc. 22nd ACM SIGKDD Int. Conf. Knowledge Discovery and Data Mining, San Francisco, CA, USA, 2016, pp. 785–794.

[11] N. V. Chawla, K. W. Bowyer, L. O. Hall, and W. P. Kegelmeyer, “SMOTE: Synthetic minority over-sampling technique,” Journal of Artificial Intelligence Research, vol. 16, pp. 321–357, 2002.

[12] M. T. Ribeiro, S. Singh, and C. Guestrin, “Why should I trust you? Ex-plaining the predictions of any classifier,” in Proc. 22nd ACM SIGKDD Int. Conf. Knowledge Discovery and Data Mining, San Francisco, CA, USA, 2016, pp. 1135–1144.

[13] J. Aguero-Valverde, “Full Bayes Poisson gamma, Poisson lognormal, and zero inflated random effects models: Comparing the precision of crash frequency estimates,” Accident Analysis & Prevention, vol. 50, pp. 289–297, 2013.

[14] A. Theofilatos and G. Yannis, “A review of the effect of traffic and weather characteristics on road safety,” Accident Analysis & Prevention, vol. 72, pp. 244–256, 2014.

[15] J. Lee, B. Abdel-Aty, and I. Yue, “Development of zone system for macro-level traffic safety analysis,” Journal of Transport Geography, vol. 25, pp. 13–21, 2021.

[16] S. Moosavi, M. H. Samavatian, S. Parthasarathy, and R. Ramnath, “A countrywide traffic accident dataset,” arXiv preprint, arXiv:1906.05409, 2019.

[17] Department for Transport, “Road Safety Data,” UK Government Open Data, 2022. [Online]. Available: https://data.gov.uk/dataset/ cb7ae6f0-4be6-4935-9277-47e5ce24a11f/road-safety-data

[18] S. Hochreiter and J. Schmidhuber, “Long short-term memory,” Neural Computation, vol. 9, no. 8, pp. 1735–1780, 1997.

[19] F. Pedregosa et al., “Scikit-learn: Machine learning in Python,” Journal of Machine Learning Research, vol. 12, pp. 2825–2830, 2011.

[20] S. M. Mohammad and P. D. Turney, “Crowdsourcing a word–emotion association lexicon,” Computational Intelligence, vol. 29, no. 3, pp. 436–465, 2013.

@article{1718000,
      author = {Kavinesh V S, Prof. Rakshitha B S},
      title = {ML-based Road Accident Hotspot Detection},
      journal = {Iconic Research And Engineering Journals},
      year = {2026},
      volume = {9},
      number = {11},
      pages = {2980-2987},
      issn = {2456-8880},
      url = {https://www.irejournals.com/formatedpaper/1718000.pdf},
      abstract = {Road traffic accidents remain one of the leading causes of fatalities and injuries worldwide, imposing significant social, economic, and public health burdens. Identifying accident-prone locations, commonly referred to as hotspots, is a critical step toward reducing road casualties through targeted safety interventions. However, traditional statistical and rule-based methods struggle to capture the complex, non-linear relationships among the numerous factors contributing to road accidents.
This paper presents a comprehensive review of machine learning methods applied to road accident hotspot detection. We survey existing approaches spanning statistical models, kernel density estimation, supervised classifiers, ensemble methods, deep learning architectures, and explainable AI techniques. We com-pare their reported performance, datasets used, feature spaces ex-plored, and key limitations. The review identifies critical research gaps, including the lack of systematic multi-model comparisons on balanced hotspot datasets, insufficient use of explainability techniques such as SHAP, and limited integration of diverse feature groups spanning spatial, temporal, road geometry, and weather attributes. The findings of this review provide a structured foundation for future research toward accurate, efficient, and interpretable road accident hotspot detection systems.},
      keywords = {Road Accident Hotspot Detection, Machine Learning, Random Forest, XGBoost, SHAP, Traffic Safety, Spa-tial Analysis, GIS, Predictive Modelling, Explainable AI},
      month = {May},
      doi = {https://doi.org/10.64388/IREV9I11-1718000}
  }