Testing The Performance of Random Forest and Support Vector Machine Algorithms for Predicting Cyclist Casualty Severity

Öz

Traditional statistical regression models for predicting casualty severity have fundamental limitations. Machine learning algorithms for classifications have started to be applied in severity analysis in order to relax the assumptions and provide better accuracy in the models. However, the performances of highly advised classification algorithms for predicting cyclist casualty severity, which particularly occurred at roundabouts, have not been investigated comprehensively. Therefore, the study in this paper developed classification models for cyclist casualty severity prediction by applying the highest two advised algorithms in the literature namely Random Forest and Support Vector Machine. The dataset included 439 cyclist casualties which were recorded at give-way roundabouts in the North East of England. The predictive variables were sociodemographic information about cyclists, weather conditions, behavior-related contributory factors, speed limit, and roundabout geometrical parameters. 70% of the records were randomly selected for the training stage and 30% were used for the testing in both Random Forest and Support Vector Machine algorithms. After training the algorithm, the testing results showed that the Random Forest algorithm predicted the outcomes with 88.6% classification accuracy. On the other hand, Support Vector Machine algorithm predicted the testing values with 84.73% classification accuracy. The algorithms misestimated 18 and 20 of the casualties in Random Forest and Support Vector Machine, respectively. The outcomes suggested that both Random Forest and Support Vector Machine algorithms were applicable for cyclist casualty severity prediction models with high performance.

Anahtar Kelimeler

• Machine learning
• Cyclist safety
• Geometry
• Roundabout
• Classification accuracy

Bisikletli Yaralanma Derecesini Tahmin Etmek için Kullanılan Random Forest ve Support Vector Machine Algoritmalarının Performanslarının Test Edilmesi

Öz

Kaza yaralanma derecesinin tahmininde kullanılan geleneksel istatistiksel regresyon modellerinin birtakım kısıtlamaları vardır. Varsayımları ortadan kaldırmak ve modellerde daha iyi doğruluk sağlamak amacıyla makine öğrenmesi tekniğinin sınıflandırma algoritmaları yaralanma derecesi analizinde uygulanmaya başlanmıştır. Ancak, uygulanması tavsiye edilen sınıflandırma algoritmalarının performansları, özellikle dönel kavşaklarda meydana gelen bisikletli kazalarının yaralanma derecesini tahmin etmek için, kapsamlı bir şekilde araştırılmamıştır. Bu sebeple, bu makaledeki çalışmada, literatürde en sık önerilen iki algoritma olan Random Forest ve Support Vector Machine'i bisikletli yaralanma derecesinin tahmininde kullanarak sınıflandırma modelleri geliştirmiştir. Veri seti, İngiltere'nin kuzey-doğu bölgesinde karma kullanımlı trafiğe sahip dönel kavşaklarda meydana gelen 439 bisikletli kazalarını içermektedir. Bağımsız değişkenler bisikletlilerin sosyodemografik bilgileri, hava koşulları, sürücü davranışı ile ilgili faktörler, hız limiti ve kavşak geometrik parametreleridir. Hem Random Forest hem de Support Vector Machine algoritmalarının eğitim aşamasında veri setinin %70’i, test aşamasında ise %30’u kullanılmıştır. Algoritmaların test aşamasından sonra ortaya çıkan sonuçlara göre, Random Forest yönteminin sınıflandırma doğruluğunun %88.6 olduğu belirlenmiştir. Support Vector Machine algoritmasının ise %84.73 sınıflandırma doğruluğu ile tahmin modeli oluşturduğu tespit edilmiştir. Yanlış tahmin edilen veri sayısı Random Forest yönteminde 18 iken Suppport Vector Machine yönteminde 20’dir. Sonuçlar, hem Random Forest hem de Support Vector Machine algoritmalarının, bisikletli kaza yaralanma derecesi tahmin modelleri oluşturmak için yüksek performansa sahip uygulanabilirliklerinin olduğunu göstermektedir.

Anahtar Kelimeler

• Makine öğrenmesi
• Bisikletli güvenliği
• Geometri
• Dönel kavşak
• Sınıflandırma doğruluğu

Kaynakça

1. [1] Silvano AP, Ma X, Koutsopoulos HN. “When do drivers yield to cyclists at unsignalized roundabouts”. Transportation Research Record: Journal of the Transportation Research Board, 2520, 2015.
2. [2] Silvano AP, Linder, A. “Traffic safety for cyclists in roundabouts: Geometry, traffic, and priority rules”. Swedish National Road and Transport Research Institute, 2017.
3. [3] Bruce W, Rodegerdts L, Scarborough W, Kittelson W, Troutbeck R, Brilon W, Bondzio L, Courage K, Kyte M, Mason J, Flannery A, Myers E, Bunker J, Jacquemart G. “Roundabouts: an informational guide”. US Department of Transport: Federal Highway Administration, AASHTO, 2000.
4. [4] Poudel N, Singleton PA. “Bicycle safety at roundabouts: a systematic literature review”. Transport Reviews, 41 (5), 617-642, 2021.
5. [5] Retting RA, Persaud BN, Garder PE, Lord D. “Crash and injury reduction following 17 installation of roundabouts in the United States”. American Journal of Public Health, 91 (4), 628-31, 2001.
6. [6] Gross F, Lyon C, Persaud B, Srinivasan R. “Safety effectiveness of converting signalized intersections to roundabouts”. Accident Analysis & Prevention, 50, 234–241, 2013.
7. [7] De Brabander B, Vereeck L. “Safety effects of roundabouts in Flanders: signal type, speed limits and vulnerable road users”. Accident Analysis & Prevention, 39 (3), 591-599, 2007.
8. [8] Furtado G. “Accommodating vulnerable road users in roundabout design”. Annual Conference of the Transportation, Canada, Quebec City, 2004.

9. [9] Daniels S, Brijs T, Nuyts E, Wets G. “Injury crashes with bicyclists at roundabouts: influence of some location characteristics and the design of cycle facilities”. Journal of Safety Research, 40 (2), 141-148, 2009.
10. [10] Jensen SU. “Safe roundabouts for cyclists”. Accident Analysis & Prevention, 105, 30-37, 2017.
11. [11] Robinson BW, Rodegerdts L, Scarborough W, Kittelson W, Troutbeck R, Brilon W, Bondzio L, Courage K, Kyte M, Mason J, Flannery A, Myers E, Bunker J, Jacquemart G. “Roundabouts: an informational guide”. FHWA-RD-00-067, Project 2425, Informational Guide Book, 2000.
12. [12] Persaud BN, Retting RA, Garder PE, Lord D. “Observational before-after study of the safety effect of U.S. roundabout conversions using the empirical Bayes method”. Annual Meeting of the Transportation Research Board, TRB ID: 01-0562, 2001.
13. [13] Elvik R. “Effects on road safety of converting intersections to roundabouts: review of evidence from non-US studies”. Transportation Research Record: Journal of the Transportation Research Board, 1847 (1), 2003.
14. [14] Arnold LS, Flannery A, Ledbetter L, Bills T, Jones MG, Ragland DR, Spautz L. “Identifying factors that determine bicyclist and pedestrian: involved collision rates and bicyclist and pedestrian demand at multi-lane roundabouts”. UC Berkeley Safe Transportation Research & Education Center, I.o.T.S. University of California, Berkeley, 2010.
15. [15] Davies DG, Taylor MC, Ryley TJ, Halliday ME. “Cyclists at roundabouts — the effects of ‘Continental’ design on predicted safety and capacity”. Transport Research Laboratory, 1997.
16. [16] Hels T, Orozova-Bekkevold I. “The effect of roundabout design features on cyclist accident rate”. Accident Analysis & Prevention, 39 (2), 300-307, 2007.
17. [17] Montella A. “Identifying crash contributory factors at urban roundabouts and using association rules to explore their relationships to different crash types”. Accident Analysis & Prevention, 43 (4), 1451-1463, 2011.
18. [18] Akgün N, Dissanayake D, Thorpe N, Bell MC. “Cyclist casualty severity at roundabouts – To what extent do the geometric characteristics of roundabouts play a part?”. Journal of Safety Research, 67, 83–91, 2018.
19. [19] Akgün N, Daniels S, Bell MC, Nuyttens N, Thorpe N, Dissanayake D. “Exploring regional differences in cyclist safety at roundabouts: A comparative study between the UK (based on Northumbria data) and Belgium”. Accident Analysis & Prevention, 150, 105902, 2021.
20. [20] Wang C, Quddus MA, Ison SG. “Predicting crash frequency at their severity levels and its application in site ranking using a two-stage mixed multivariate model”. Accident Analysis & Prevention, 43 (6) (2011), 1979-1990, 2011.
21. [21] Savolainen P, Mannering F, Lord D, Quddus MA. “The statistical analysis of crash-injury severities: a review and assessment of methodological alternatives”. Accident Analysis & Prevention, 43 (5) (2011), 1666-1676, 2011.
22. [22] Daniels S, Brijs T, Nuyts E, Wets G. “Externality of risk and crash severity at roundabouts”. Accident Analysis & Prevention, 42 (6), 1966-1973, 2010.
23. [23] Daniels S, Brijs T, Nuyts E, Wets G. “Extended prediction models for crashes at roundabouts”. Safety Science, 49 (2), 198-207, 2011.
24. [24] Møller M, Hels T. “Cyclists’ perception of risk in roundabouts”. Accident Analysis & Prevention, 40 (3), 1055-1062, 2008.
25. [25] Field A. Discovering Statistics Using SPSS. 3rd Edition, SAGE Publications Ltd, 2009.
26. [26] Peduzzi P, Concato J, Kemper E, Holford TR, Feinstein AR. “A simulation study of the number of events per variable in logistic regression analysis”. Journal of Clinical Epidemiology, 49 (12), 1373-1379, 1996.
27. [27] Silva PB, Andrade M, Ferreira S. “Machine learning applied to road safety modeling: A systematic literature review”. Journal of Traffic and Transportation Engineering (English Edition), 7 (6), 775-790, 2020.
28. [28] Santos K, Dias JP, Amado C. “A literature review of machine learning algorithms for crash injury severity prediction”. Journal of Safety Research, 80, 254-269, 2022.
29. [29] Janstrup KH, Kostic B, Møller M, Rodrigues F, Borysov S, Pereira FC. “Predicting injury-severity for cyclist crashes using natural language processing and neural network modelling”. Safety Science, 164, 106153, 2023.
30. [30] Katanalp BY, Eren E. “The novel approaches to classify cyclist accident injury-severity: Hybrid fuzzy decision mechanisms”. Accident Analysis & Prevention, 144, 105590, 2020.
31. [31] Vilaça M, Macedo E, Coelho MC. “A Rare event modelling approach to assess injury severity risk of vulnerable road users”. Safety, 5(2), 29, 2019.
32. [32] Yamparala R, Challa R, Valeti P, Chaitanya PS. "Prediction of cyclist road accidents in india using machine learning and visualization techniques". Second International Conference on Artificial Intelligence and Smart Energy (ICAIS), Coimbatore, India, 476-481, 2022.
33. [33] Zhang Y, Li H, Ren G. “Analyzing the injury severity in single-bicycle crashes: An application of the ordered forest with some practical guidance”. Accident Analysis & Prevention, 189, 107126, 2023.
34. [34] Birfir S, Elalouf A, Rosenbloom T. “Building machine-learning models for reducing the severity of bicyclist road traffic injuries”. Transportation Engineering, 12, 100179, 2023.
35. [35] Cara I, De Gelder E. "Classification for safety-critical car-cyclist scenarios using machine learning". IEEE 18th International Conference on Intelligent Transportation Systems, Gran Canaria, Spain, 1995-2000, 2015.
36. [36] Fan J, Ma C, Zhong Y. “A selective overview of deep learning”. Statistical Science, 36(2), 264-290, 2021
37. [37] Li L, Shrestha S, Hu G. “Analysis of road traffic fatal accidents using data mining techniques”. Ieee 15th International Conference on Software Engineering Research, Management and Applications (Sera), 363–70, 2021.
38. [38] Boser BE, Guyon IM, Vapnik VN. "A training algorithm for optimal margin classifiers". Proceedings of the 5th Annual Workshop on Computational Learning Theory, 144-152, 1992.
39. [39] Suthaharan S. “Support vector machine - In Machine learning models and algorithms for big data classification”. Integrated Series in Information Systems, 207–235, Springer, 2016.
40. [40] Wahab L. Jiang H. “A comparative study on machine learning based algorithms for prediction of motorcycle crash severity”. PLoS one, 14-4, 2019.
41. [41] Zhang J, Li Z, Pu Z, Xu C. "Comparing Prediction Performance for Crash Injury Severity Among Various Machine Learning and Statistical Methods". IEEE Access, 6, 60079-60087, 2018.
42. [42] DfT. “TD 16-07 Geometric design of roundabouts”. Department for Transport, the UK, 2007.