Spatial-Temporal CNNs: A Deep Learning Approach to Predictive Traffic Accident Hotspotting
DOI:
https://doi.org/10.70882/noun-ijcea.2026.1114Keywords:
Accident, CNN, Deep Learning, DistillBERT, Road Traffic, Special pattern, TransformerAbstract
Road traffic accidents remain a serious global challenge, claiming millions of lives annually and straining economies worldwide. This study introduces a novel hybrid architecture that integrates Convolutional Neural Networks with DistillBERT transformer-based architecture to tackle the persistent problem of accident severity prediction. However, by merging different real-world data streams weather patterns, traffic congestion levels, road infrastructure details, and historical crash records from Los Angeles County. The model learns to read the complex language of urban risk. The approach transforms raw, noisy, and imbalanced data into meaningful structure sequence, enabling the CNN to map spatial pattern while DistillBERT's lightweight attention mechanism traces how these risks evolve over time. The experimental training and evaluation show that CNN-DistillBERT’s models achieve the highest results of 94.66% accuracy in predicting accident severity. This study opens practical approach for real time road traffic analysis before and after it happened.
References
Ahmed, S., Hossain, M. A., Ray, S. K., Bhuiyan, M. M. I., & Sabuj, S. R. (2023). A study on road accident prediction and contributing factors using explainable machine learning models: Analysis and performance. Transportation Research Interdisciplinary Perspectives, 19, 100814. https://doi.org/10.1016/j.trip.2023.100814
Ahmed, S. K., Mohammed, M. G., Abdulqadir, S. O., El-Kader, R. G. A., El-Shall, N. A., Chandran, D., Rehman, M. E. U., & Dhama, K. (2023). Road traffic accidental injuries and deaths: A neglected global health issue. Health Science Reports, 6(5), e1240. https://doi.org/10.1002/hsr2.1240
Alaei, S. (2023). An Automated Discharge Summary System Built for Multiple Clinical English Texts by Pre-trained DistilBART Model. https://urn.kb.se/resolve?urn=urn:nbn:se:su:diva-222616
Alharbi, K., & Haq, M. A. (2024). Enhancing Disaster Response and Public Safety with Advanced Social Media Analytics and Natural Language Processing. Engineering, Technology & Applied Science Research, 14(3), 14212–14218. https://doi.org/10.48084/etasr.7232
Azhar, A., Rubab, S., Khan, M. M., Bangash, Y. A., Alshehri, M. D., Illahi, F., & Bashir, A. K. (2023). Detection and prediction of traffic accidents using deep learning techniques. Cluster Computing, 26(1), 477–493. https://doi.org/10.1007/s10586-021-03502-1
Berhanu, Y., Alemayehu, E., & Schröder, D. (2023). Examining Car Accident Prediction Techniques and Road Traffic Congestion: A Comparative Analysis of Road Safety and Prevention of World Challenges in Low-Income and High-Income Countries. Journal of Advanced Transportation, 2023, 1–18. https://doi.org/10.1155/2023/6643412
Boukerche, A., & Wang, J. (2020). Machine Learning-based traffic prediction models for Intelligent Transportation Systems. Computer Networks, 181, 107530. https://doi.org/10.1016/j.comnet.2020.107530
Ekanem, I. (2025). Analysis of Road Traffic Accident Using AI Techniques. Open Journal of Safety Science and Technology, 15(1), Article 1. https://doi.org/10.4236/ojsst.2025.151004
Finogeev, A., Deev, M., Finogeev, A., & Kolesnikoff, I. (2020). Proactive Big Data Analysis for Traffic Accident Prediction1. 2020 5th International Conference on Innovative Technologies in Intelligent Systems and Industrial Applications (CITISIA), 1–9. https://doi.org/10.1109/CITISIA50690.2020.9371796
Fu, X., Liu, J., Jones, S., Barnett, T., & Khattak, A. J. (2022). From the past to the future: Modeling the temporal instability of safety performance functions. Accident Analysis & Prevention, 167, 106592. https://doi.org/10.1016/j.aap.2022.106592
Jaulkar, S., & Parihar, A. (2025). Different types of injury associated with road traffic accidents. Multidisciplinary Reviews, 8(11). https://malque.pub/ojs/index.php/mr/article/view/5545
Kashinath, S. A., Mostafa, S. A., Mustapha, A., Mahdin, H., Lim, D., Mahmoud, M. A., Mohammed, M. A., Al-Rimy, B. A. S., Fudzee, M. F. M., & Yang, T. J. (2021). Review of Data Fusion Methods for Real-Time and Multi-Sensor Traffic Flow Analysis. IEEE Access, 9, 51258–51276. https://doi.org/10.1109/ACCESS.2021.3069770
Mahmoud, A., & Mohammed, A. (2024). Leveraging Hybrid Deep Learning Models for Enhanced Multivariate Time Series Forecasting. Neural Processing Letters, 56(5), 223. https://doi.org/10.1007/s11063-024-11656-3
Manzoor, M., Umer, M., Sadiq, S., Ishaq, A., Ullah, S., Madni, H. A., & Bisogni, C. (2021). RFCNN: Traffic Accident Severity Prediction Based on Decision Level Fusion of Machine and Deep Learning Model. IEEE Access, 9, 128359–128371. https://doi.org/10.1109/ACCESS.2021.3112546
Rahim, M. A., & Hassan, H. M. (2021). A deep learning based traffic crash severity prediction framework. Accident Analysis & Prevention, 154, 106090. https://doi.org/10.1016/j.aap.2021.106090
RajyaLakshmi, T., & Kuppusamy, K. S. (2023). Comparative Analysis of Pretrained Encoder-Decoder Transformer Models for Extreme Text Summarization. 2023 Second International Conference on Advances in Computational Intelligence and Communication (ICACIC), 1–6. https://doi.org/10.1109/ICACIC59454.2023.10435363
Sajadi, P., Qorbani, M., Moosavi, S., & Hassannayebi, E. (2024). Accident Impact Prediction based on a deep convolutional and recurrent neural network model (arXiv:2411.07537). arXiv. https://doi.org/10.48550/arXiv.2411.07537
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Dodo, Simon Numduck, Dr. Muhammad, Kudu Muhammad, Dr. Sulaimon, Adebayo Bashir, Dr. Enesi, Femi Aminu (Author)
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
