Türkiye'de 2021-2024 Yılları Arasında Makine Öğrenmesi Modelleri Kullanılarak Trafik Kazası Analizi ve Tahmini

Abstract

Trafik kazaları, kamu güvenliği ve kentsel gelişim açısından önemli bir sorun teşkil etmektedir. Son yıllarda, dünya nüfusunun artması ve araç sayısının çoğalması nedeniyle trafik kazalarının sayısı yükselmiş, bu da çok sayıda ölüm ve yaralanmaya yol açmıştır. Bu çalışma, 2021-2024 yılları arasında Türkiye'nin beş büyük şehrindeki trafik kazalarını inceleyerek eğilimleri belirlemeyi ve doğrusal regresyon ile rastgele orman regresyon modelleri kullanarak gelecekteki kazaları tahmin etmeyi amaçlamaktadır. Bu analiz için veriler, Türkiye Cumhuriyeti İçişleri Bakanlığı Jandarma Genel Komutanlığı'ndan temin edilmiştir. Model performansını değerlendirmek için Ortalama Mutlak Hata, Ortalama Kare Hatası ve R-kare gibi temel değerlendirme ölçütleri kullanılmıştır. Sonuçlar, kazaların şehirler, aylar ve yıllar arasında önemli farklılıklar gösterdiğini ortaya koymaktadır. Ayrıca bulgular, makine öğrenimi modellerinin trafik kazalarını yüksek doğrulukla tahmin etme konusundaki etkinliğini vurgulamaktadır. İki model arasında, rastgele orman regresörünün, değerlendirme ölçütleri açısından doğrusal regresyondan daha üstün performans gösterdiği belirlenmiştir. Bunun yanı sıra, analiz sonuçları beş şehirde, özellikle Ankara ve İzmir'de, kaza, ölüm ve yaralanma sayılarında artış eğilimi olduğunu göstermektedir. Bu öngörüsel ve analitik bulgular, trafik kazalarını azaltmak ve yol güvenliğini artırmak için politika yapıcılar ve araştırmacılar için yol gösterici olabilir.

Keywords

• Doğrusal regresyon
• Makine öğrenimi
• Rastgele Orman
• Trafik kazası

Traffic Accident Analysis and Prediction Using Machine Learning Models in Türkiye from 2021 to 2024

Abstract

Traffic accidents represent a major challenge to public safety and urban development. In recent years, the number of road traffic accidents has been increasing due to the rising global population and the growing number of vehicles, leading to numerous fatalities and injuries. This study examines traffic accidents in five major cities of Türkiye from 2021 to 2024, aiming to identify trends and predict future accidents using linear regression and random forest regressor models. Data for this analysis were obtained from the Gendarmerie General Command of the Ministry of Internal Affairs, Republic of Türkiye. To evaluate model performance, key metrics such as Mean Absolute Error, Mean Squared Error, and R-squared were utilized. The results indicate significant variations in accident patterns across cities, months, and years. Furthermore, findings highlight the effectiveness of machine learning models in predicting traffic incidents with high accuracy. Among the two models, the random forest regressor outperforms linear regression in terms of evaluation metrics. Moreover, the analytical results indicate an upward trend in accidents, fatalities, and injuries across the five cities, particularly in Ankara and İzmir. These predictive and analytical insights can provide valuable guidance for policymakers and researchers in formulating effective strategies to mitigate traffic accidents and enhance road safety.

Keywords

• Linear regression
• Machine learning
• Random Forest
• Traffic accident

References

1. Aygencel, G., Karamercan, M., Ergin, M. & Telatar, G. (2008). Review of traffic accident cases presenting to an adult emergency service in Turkey. Journal of Forensic and Legal Medicine, 15(1), 1-6. https://doi.org.10.1016/j.jflm.2007.05.005.
2. World Health Organization. (2023). Global status report on road safety 2023: Summary (Licence: CC BY-NC-SA 3.0 IGO). World Health Organization. https://www.who.int/teams/social-determinants-of-health/safety-and-mobility/global-status-report-on-road-safety-2023.
3. Turkish Statistical Institute. (2024). Road traffic accident statistics, 2023 (Issue 53479). https://data.tuik.gov.tr/Bulten/Index?p=Road-Traffic-Accident-Statistics-2023-53479&dil=2
4. Kuyumcu, Z. C., Aslan, H., & Yurtay, N. (2024). Casualty analysis of the drivers in traffic accidents in Turkey: A CHAID decision tree model. Applied Sciences, 14(24), 11693. https://doi.org/10.3390/app142411693
5. Gendarmerie General Command. (2025, January). Aylık istatistik bültenleri (Monthly statistical bulletins). https://www.jandarma.gov.tr/veriler.
6. Erdogan, S. (2009). Explorative spatial analysis of traffic accident statistics and road mortality among the provinces of Turkey. Journal of Safety Research, 40(5), 341-351. https://doi.org.10.1016/j.jsr.2009.07.006.
7. Celik, A. K., & Oktay, E. (2014). A multinomial logit analysis of risk factors influencing road traffic injury severities in the Erzurum and Kars Provinces of Turkey. Accident Analysis & Prevention, 72, 66–77. https://doi.org/10.1016/j.aap.2014.06.010
8. Sungur, İ., Akdur, R., & Piyal, B. (2014). Analysis of traffic accidents in Turkey, Ankara Medical Journal, 14(3), 114-124.

9. Kaygisiz, Ö., Senbil, M., & Yildiz, A. (2017). Influence of urban built environment on traffic accidents: The case of Eskisehir (Turkey). Case Studies on Transport Policy, 5(2), 306-313. https://doi.org.10.1016/j.cstp.2017.02.002.
10. Ihueze, C. C., & Onwurah, U. O. (2018). Road traffic accidents prediction modelling: An analysis of Anambra State, Nigeria. Accident Analysis & Prevention, 112, 21-29. https://doi.org.10.1016/j.aap.2017.12.016.
11. Özen, M. (2018). Comparative study of regional crash data in Turkey. Turkish Journal of Engineering, 2(3), 113-118. https://doi.org.10.31127/tuje.385008.
12. Kumeda, B., Zhang, F., Zhou, F., Hussain, S., Almasri, A., & Assefa, M. (2019). Classification of road traffic accident data using machine learning algorithms. IEEE 11th International Conference on Communication Software and Networks, 682-687, https://doi.org.10.1109/ICCSN.2019.8905362.
13. Erenler, A. K., & Gumus, B. (2019). Analysis of road traffic accidents in Turkey between 2013 and 2017. Medicina (Kaunas), 55(10), 679. https://doi.org.10.3390/medicina55100679.
14. Al Mamlook, R. E., Ali, A., Hasan, R. A., & Kazim, H. A. M. (2019). Machine learning to predict the freeway traffic accidents-based driving simulation. IEEE National Aerospace and Electronics Conference, 630-634. https://doi.org.10.1109/NAECON46414.2019.9058268.
15. Labib, M. F., Rifat, A. S., Hossain, M. M., Kumar, A., & Nawrine, F. (2019). Road accident analysis and prediction of accident severity by using machine learning in Bangladesh. 7th International Conference on Smart Computing & Communications, 1-5. https://doi.org.10.1109/ICSCC.2019.8843640.
16. Qu, Y., Lin, Z., Li, H., & Zhang, X. (2019). Feature recognition of urban road traffic accidents based on GA-XGBoost in the context of big data. IEEE Access, 7, 170106-170115, https://doi.org.10.1109/access.2019.2952655.
17. AlKheder, S., AlRukaibi, F., & Aiash, A. (2020). Risk analysis of traffic accidents' severities: An application of three data mining models. ISA Transactions, 106, 213-220. https://doi.org.10.1016/j.isatra.2020.06.018.
18. Yassin, S. S., & Pooja, R. (2020). Road accident prediction and model interpretation using a hybrid K-means and random forest algorithm approach. SN Applied Sciences, 2(9), Article 1560. https://doi.org.10.1007/s42452-020-3125-1.
19. Chen M. M., & Chen, M. C. (2020). Modeling road accident severity with comparisons of logistic regression, decision tree and random forest. Information, 11(5), 270. https://doi.org.10.3390/info11050270.
20. Sangare, M., Gupta, S., Bouzefrane, S., Banerjee, S., & Muhlethaler, P. (2021). Exploring the forecasting approach for road accidents: Analytical measures with hybrid machine learning. Expert Systems with Applications, 167, 113855. https://doi.org.10.1016/j.eswa.2020.113855.
21. Bokaba, T., Doorsamy, W., & Paul, B. S. (2022). Comparative study of machine learning classifiers for modelling road traffic accidents. Applied Sciences, 12(2), 828. https://doi.org.10.3390/app12020828.
22. Korkmaz, A. (2023). Predictive modeling of urban traffic accident severity in Türkiye's centennial: machine learning approaches for sustainable cities. Kent Akademisi, 16, 395-406. https://doi.org.10.35674/kent.1353402.
23. Segal, M. R. (2004). Machine learning benchmarks and random forest regression. UCSF: Center for Bioinformatics and Molecular Biostatistics, https://escholarship.org/uc/item/35x3v9t4.
24. Breiman, L. (2001). Random Forests. Machine Learning, 45, 5-32. https://doi.org/10.1023/A:1010933404324.
25. Poole M. A., & O'Farrell, P. N. (1971). The assumptions of the linear regression mode. Transactions of the Institute of British Geographers, 52, 145-158. https://doi.org/10.2307/621706.
26. Aalen, O. O. (1989). A linear regression model for the analysis of life times. Stat Med, 8(8), 907-925. https://doi.org.10.1002/sim.4780080803.
27. Twomey, J. M., & Smith, A. E. (1995). Performance measures, consistency, and power for artificial neural network models. Mathematical and Computer Modelling, 21(1-2), 243-258.
28. Serefoglu Cabuk, K., Cengiz, S. K., Guler, M. G., Topcu, H., Cetin Efe, A., Ulas, M. G., & Poslu Kandemir, F. (2024). Chasing the objective upper eyelid symmetry formula; R2, RMSE, POC, MAE, and MSE. International Ophthalmology, 44(1), 303, https://doi.org.10.1007/s10792-024-03157-y.