Küresel Yol Güvenliğini Geliştirmek İçin Makine Öğreniminden Yararlanma: Kapsamlı Bir İnceleme

Öz

Küresel kentleşme hızlanırken, yol güvenliği, artan trafik kazaları ve ölümlerin altını çizdiği acil bir endişe olmaya devam etmektedir. Karayolu Trafik Yaralanmaları (RTI), dünya çapında sekizinci önde gelen ölüm nedeni haline geldi. Makale, trafik kazalarını, bunların ciddiyetini ve nedensel faktörleri tahmin etmede makine öğreniminin potansiyelini derinlemesine araştırmaktadır. Bu çalışma, Addis Ababa Şehri Polis Departmanından alınan trafik kazası kayıtları üzerindeki makine öğrenimi modellerini kapsamlı bir şekilde değerlendirmektedir. 15 özelliğe sahip 12.316 kayıttan oluşan veri setinde, Sentetik Azınlık Aşırı Örnekleme Tekniği (SMOTE) ve Min-Max ölçekleme başta olmak üzere ön işleme teknikleri uygulanmıştır. Beş algoritma – Random Forest (RF), Gaussian Naive Bayes, CatBoostClassifier, LightGBM ve XGBoost – tahmin doğruluğu açısından test edilmiştir. Bulgular, SMOTE ve Min-Max uygulamasından sonra %92,2'lik bir tepe doğruluğu elde eden RF modelinin hakimiyetine ışık tutmaktadır. Mevcut literatürle karşılaştırmalı bir analiz, RF'nin çeşitli veri kümelerinde yinelenen etkili bir model olmasına rağmen, veri ön işlemenin ve belirli veri kümelerine model uygunluğunun öneminin çok önemli olduğunu göstermiştir. Bu çalışma, trafik kazası analizinde makine öğreniminin potansiyelinin ve araştırmacıların optimum sonuçlar için yapması gereken incelikli seçimlerin altını çizmektedir.

Anahtar Kelimeler

• Trafik Kazası Analizi
• Makine Öğrenmesi
• Rastgele Orman
• Min-Maks Ölçeklendirme
• Karşılaştırmalı Çalışma.

Harnessing Machine Learning to Enhance Global Road Safety: A Comprehensive Review

Öz

As global urbanization accelerates, road safety remains a pressing concern, underscored by escalating traffic accidents and fatalities. Road Traffic Injuries (RTI) have become the eighth leading cause of death worldwide. The article delves deep into the potential of machine learning in predicting traffic accidents, their severity, and causal factors. This study comprehensively evaluates machine learning models on traffic accident records sourced from the Addis Ababa City Police Department. Comprising 12,316 records with 15 features, the dataset underwent preprocessing techniques, specifically Synthetic Minority Over-sampling Technique (SMOTE) and Min-Max scaling. Five algorithms – Random Forest (RF), Gaussian Naive Bayes, CatBoostClassifier, LightGBM, and XGBoost – were tested for their prediction accuracy. The findings spotlight the dominance of the RF model, achieving a peak accuracy of 92.2% post-SMOTE and Min-Max application. A comparative analysis with existing literature showed that while RF is a recurrently effective model across various datasets, data preprocessing and model suitability to specific datasets is paramount. This study underscores the potential of machine learning in traffic accident analysis and the nuanced choices researchers must make for optimal outcomes.

Anahtar Kelimeler

• Traffic Accident Analysis
• Machine Learning
• Random Forest
• Min-Max Scaling
• Comparative Study.

Kaynakça

1. [1] Fallon I. and O’Neill D., “The world’s first automobile fatality,” Accid. Anal. Prev., vol. 37, no. 4, pp. 601–603, (2005).
2. [2] “When did the first motoring fatality occur?,” National Motor Museum, 11-Jan-2018. [Online]. Available: https://nationalmotormuseum.org.uk/ufaqs/when-did-the-first-motoring-fatality-occur/. [Accessed: 21-Aug-2023].
3. [3] ”2022'de Türkiye'de artan trafik kazası sayısı.”, Atlas Magazine, 31-April-2023. [Online]. Available: https://www.atlas-mag.net/en/category/pays/turquie/rising-number-of-road-accidents-in-turkey-in-2022#:~:text=The%20Turkish%20Statistical%20Institute%20(TurkStat,the%20remainder%20in%20material%20damage. [Accessed: 21-Aug-2023].
4. [4] Thelancet.com. [Online]. Available: https://www.thelancet.com/infographics-do/road-safety-2022. [Accessed: 21-Aug-2023].
5. [5] “Global status report on road safety 2018,” Whoint, 17-Jun-2018. [Online]. Available: https://www.who.int/publications/i/item/9789241565684. [Accessed: 21-Aug-2023].
6. [6] Li L., Zhu L., and Sui D. Z., “A GIS-based Bayesian approach for analyzing spatial–temporal patterns of intra-city motor vehicle crashes,” J. Transp. Geogr., vol. 15, no. 4, pp. 274–285, (2007).
7. [7] Tola A. M., Demissie T. A., Saathoff F., and Gebissa A., “Severity, spatial pattern and statistical analysis of road traffic crash hot spots in Ethiopia,” Appl. Sci. (Basel), vol. 11, no. 19, p. 8828, (2021).
8. [8] “3 star or better,” iRAP, 02-Aug-2017. [Online]. Available: https://irap.org/3-star-or-better/. [Accessed: 21-Aug-2023].

9. [9] Gutierrez-Osorio C., González F. A., and Pedraza C. A., “Deep Learning ensemble model for the prediction of traffic accidents using social media data,” Computers, 11(9): 126, (2022).
10. [10] “List of countries by traffic-related death rate” Wikipedia. 12-Aug-2023. [Online]. Available: https://en.wikipedia.org/wiki/List_of_countries_by_traffic-related_death_rate . [Accessed: 21-Aug-2023].
11. [11] Archive.org. [Online]. Available: https://web.archive.org/web/20151020144338/http://www.who.int/violence_injury_prevention/road_safety_status/2015/en/. [Accessed: 21-Aug-2023].
12. [12] Bedane T. T., Assefa B. G., and Mohapatra S. K., “Preventing traffic accidents through machine learning predictive models,” in 2021 International Conference on Information and Communication Technology for Development for Africa (ICT4DA), (2021).
13. [13] Mackay M., “Global priorities for vehicle safety,” Traffic Inj. Prev., 4(1): 1–4, (2003).
14. [14] Raja K., Kaliyaperumal K., Velmurugan L., and Thanappan S., “Forecasting road traffic accident using deep artificial neural network approach in case of Oromia Special Zone,” Soft Comput., (2023).
15. [15] Beshah T.,, and Hill S., “Mining road traffic accident data to improve safety: role of road-related factors on accident severity in Ethiopia”. In the 2010 AAAI Spring Symposium series, (2010).
16. [16] Chen C., Zhang G., Qian Z., Tarefder R. A., and Tian Z., “Investigating driver injury severity patterns in rollover crashes using support vector machine models,” Accid. Anal. Prev., 90: 128–139, (2016).
17. [17] Liu M., Wu J., Wang Y., and He L., “Traffic flow prediction based on deep learning.” Journal of System Simulation, 30(11): 4100, (2018).
18. [18] Zheng J. and Huang M., “Traffic flow forecast through time series analysis based on deep learning,” IEEE Access, 8: 82562–82570, (2020).
19. [19] Dong C., Shao C., Li J., and Xiong Z., “An improved deep learning model for traffic crash prediction,” J. Adv. Transp., 2018: 1–13, (2018).
20. [20] Kumeda B., Zhang F., Zhou F., Hussain S., Almasri A., and Assefa M., “Classification of road traffic accident data using machine learning algorithms,” in 2019 IEEE 11th International Conference on Communication Software and Networks (ICCSN), 2019.Networks (ICCSN), Chongqing, China, 682-687, doi: 10.1109/ICCSN.2019.8905362, (2019).
21. [21] Gan J., Li L., Zhang D., Yi Z., and Xiang Q., “An alternative method for traffic accident severity prediction: Using Deep Forests algorithm,” J. Adv. Transp., 2020: 1–13, (2020).
22. [22] Çeli̇k A. and Sevli̇ O., “Predicting traffic accident severity using machine learning techniques,” Türk Doğa ve Fen Dergisi, 11(3): 79–83, (2022).
23. [23] Ghandour A. J., Hammoud H., and Al-Hajj S., “Analyzing factors associated with fatal road crashes: A machine learning approach,” Int. J. Environ. Res. Public Health, 17(11): 4111, (2020).
24. [24] Bhuiyan, H., Ara, J., Hasib, K. M., Sourav, M. I. H., Karim, F. B., Sik-Lanyi, C., ... and Yasmin, S., “Crash severity analysis and risk factors identification based on an alternate data source: a case study of developing country,” Sci. Rep., 12(1): 21243, (2022).
25. [25] Al-Mistarehi B. W., Alomari A. H., Imam R., and. Mashaqba M, “Using machine learning models to forecast the severity level of traffic crashes by R Studio and ArcGIS”. Frontiers in the built environment, 8, 860805, (2022).
26. [26] Ahmed S., Hossain M. A., Ray S. K., Bhuiyan M. M. I., and Sabuj S. R., “A study on road accident prediction and contributing factors using explainable machine learning models: analysis and performance,” Transp. Res. Interdiscip. Perspect., 19(100814): 100814, (2023).
27. [27] Santos D., Saias J., Quaresma P., and Nogueira V. B., “Machine learning approaches to traffic accident analysis and hotspot prediction,” Computers, 10(12): 157, (2021).
28. [28] Yassin S. S. and Pooja, “Road accident prediction and model interpretation using a hybrid K-means and random forest algorithm approach,” SN Appl. Sci., 2(9): (2020).
29. [29] Bedane T. T., “Road Traffic Accident Dataset of Addis Ababa City.” Mendeley, (2020).
30. [30] Özsürünç R., “The role of data mining in digital transformation,” in Contributions to Management Science, Cham: Springer International Publishing, 177–190, (2023).
31. [31] Chawla N. V., Bowyer K. W., Hall L. O., and Kegelmeyer W. P., “SMOTE: Synthetic minority over-sampling technique,” J. Artif. Intell. Res., 16: 321–357, (2002).
32. [32] Amirruddin A. D., Muharam F. M., Ismail M. H., Tan N. P., and Ismail M. F., “Synthetic Minority Over-sampling TEchnique (SMOTE) and Logistic Model Tree (LMT)-Adaptive Boosting algorithms for classifying imbalanced datasets of nutrient and chlorophyll sufficiency levels of oil palm (Elaeis guineensis) using spectroradiometers and unmanned aerial vehicles,” Comput. Electron. Agric., 193(106646): 106646, (2022).
33. [33] Fai N. J., Wey W. K., Qi K. Y., Xian G. J., Chun R. J. M., and bin Abdul Salam Z. A., “Digits Classification Using Random Forest Classifier”. Journal of Applied Technology and Innovation (e-ISSN: 2600-7304), 7(3): 63, (2023).
34. [34] Breiman L., “Random forests”. Machine learning, 45(1): 5–32, (2001).
35. [35] Gayathri, B. M., & Sumathi, C. P. ,“An automated technique using Gaussian naïve Bayes classifier to classify breast cancer,” Int. J. Comput. Appl., 148(6): 16–21, (2016).
36. [36] Deekshitha B., Aswitha C., Sundar C. S., and. Deepthi A. K, “URL-Based Phishing Website Detection by Using Gradient and Catboost Algorithms.” Int. J. Res. Appl. Sci. Eng. Technol., 10(6): 3717–3722, (2022).
37. [37] “Welcome to LightGBM’s documentation! — LightGBM 4.0.0 documentation,” Readthedocs.io. [Online]. Available: https://lightgbm.readthedocs.io/en/stable/. [Accessed: 21-Aug-2023].
38. [38] Ramraj S.,, , UzirSunil N., R., and Banerjee S., “Experimenting XGBoost algorithm for prediction and classification of different datasets”. International Journal of Control Theory and Applications, 9(40): 651-662, (2016).
39. [39] Memon N., Patel S. B., and Patel D. P., “Comparative analysis of artificial neural network and XGBoost algorithm for PolSAR image classification,” in Lecture Notes in Computer Science, Cham: Springer International Publishing, 452–460, (2019).
40. [40] Krishnaveni S. and Hemalatha M., “A perspective analysis of traffic accident using data mining techniques,” Int. J. Comput. Appl., 23(7): 40–48, (2011).
41. [41] AlMamlook R. E., Kwayu K. M., Alkasisbeh M. R., and Frefer A. A., “Comparison of machine learning algorithms for predicting traffic accident severity,” in 2019 IEEE Jordan International Joint Conference on Electrical Engineering and Information Technology (JEEIT), (2019).
42. [42] Korkmaz, A. and Buyukgoze, S. “Detection of Fake Websites by Classification Algorithms.” Eur J. Sci. Technol., 16: 826–833, (2019).
43. [43] Rezashoar, S., Kashi, E., and Saeidi, S. “Comparison of Machine Learning Algorithms for Predicting Traffic Accident Severity (Case Study: United Kingdom from 2010 to 2014)”. doi.org/10.21203/rs.3.rs-3101818/v1. https://www.researchsquare.com/article/rs-3101818/v1. (2023).