Traffic Congestion Recognition Using Ensemble Learning with Metaheuristic OOBO-Based Hyper-Parameter Optimization

In this study, we have developed a traffic congestion recognition method based on ensemble learning and metaheuristic algorithms for overcome the sensitivity of individual ML algorithm on the dataset and hyper-parameter setting. In the proposed method, the standard dataset was collected from the open-source Kaggle repository. The dataset was pre-processed using the k-mean clustering and splitted into training and testing sets. After that, the training set was given to the five ML algorithms utilized in the ensemble learning approach and validated using the testing set. Based on their individual performance, voting classifier was utilized to determined which ML algorithm outperforms over other. Next, the selected ML algorithm parameters, hyper-tuned using the metaheuristic OOBO algorithm based on the RMSE objective function. The OOBO algorithm searches the solution space of the parameters in the lower and upper bound based on the population and iteration size. Finally, the parameter setting was done using the best values and traffic congestion recognition was done. From the simulation evaluation, we found that GB-OOBO algorithm achieves the lowest error metrics and high R-squared value over the individual ML algorithms.

[1] Y. Modi, R. Teli, A. Mehta, K. Shah, and M. Shah, “A comprehensive review on intelligent traffic management using machine learning algorithms,” Innovative Infrastructure Solutions, vol. 7, no. 1, Dec. 2021, doi: 10.1007/s41062-021-00718-3. Available: https://doi.org/10.1007/s41062-021-00718-3 [2] A. Agrahari, M. M. Dhabu, P. S. Deshpande, A. Tiwari, M. A. Baig, and A. D. Sawarkar, “Artificial Intelligence-Based Adaptive Traffic Signal Control System: A Comprehensive Review,” Electronics, vol. 13, no. 19, p. 3875, Sep. 2024, doi: 10.3390/electronics13193875. Available: https://doi.org/10.3390/electronics13193875 [3] Bakir, D., Moussaid, K., Chiba, Z. and Abghour, N., 2024, December. A comprehensive review of traffic congestion prediction models: Machine learning and statistical approaches. In 2024 IEEE International Conference on Computing (ICOCO) (pp. 539-545). IEEE. [4] R. Vatambeti et al., “Enhancing urban traffic congestion prediction through efficientnet and optimized ensemble learning models,” Scientific Reports, vol. 15, no. 1, p. 40224, Nov. 2025, doi: 10.1038/s41598-025-24012-9. Available: https://doi.org/10.1038/s41598-025-24012-9 [5] T. Zhang and S. Mao, “Machine Learning for End-to-End Congestion Control,” IEEE Communications Magazine, vol. 58, no. 6, pp. 52–57, Jun. 2020, doi: 10.1109/mcom.001.1900509. Available: https://doi.org/10.1109/mcom.001.1900509 [6] L. Hammoumi, S. Farah, M. Benayad, M. Maanan, and H. Rhinane, “Leveraging machine learning to predict traffic jams: Case study of Casablanca, Morocco,” Journal of Urban Management, vol. 14, no. 3, pp. 813–826, Mar. 2025, doi: 10.1016/j.jum.2025.02.004. Available: https://doi.org/10.1016/j.jum.2025.02.004 [7] Z. Miao and Q. Liao, “IoT-Based traffic prediction for smart cities,” IEEE Access, vol. 13, pp. 52369–52384, Jan. 2025, doi: 10.1109/access.2025.3552276. Available: https://doi.org/10.1109/access.2025.3552276 [8] S. Gupta, A. Kumar, and A. Kumar, “Analysis of Traffic Flow Congestion by Integrating Supervised Machine Learning with K-mean Clustering,” SN Computer Science, vol. 6, no. 3, Mar. 2025, doi: 10.1007/s42979-025-03761-4. Available: https://doi.org/10.1007/s42979-025-03761-4 [9] S. Alsubai, A. K. Dutta, and A. R. W. Sait, “Hybrid deep learning-based traffic congestion control in IoT environment using enhanced arithmetic optimization technique,” Alexandria Engineering Journal, vol. 105, pp. 331–340, Jul. 2024, doi: 10.1016/j.aej.2024.06.072. Available: https://doi.org/10.1016/j.aej.2024.06.072 [10] N. Deepika and G. Pandove, “A comparison of ML models for predicting congestion in urban cities,” International Journal of Intelligent Transportation Systems Research, vol. 22, no. 1, pp. 171–188, Feb. 2024, doi: 10.1007/s13177-024-00387-3. Available: https://doi.org/10.1007/s13177-024-00387-3