Journal of the Korea Society of Computer and Information (한국컴퓨터정보학회논문지)

Korean Society of Computer Information (한국컴퓨터정보학회)
권호 표지
DOI QR코드

DOI QR Code

Driver Group Clustering Technique and Risk Estimation Method for Traffic Accident Prevention

  • Tae-Wook Kim (Division of Software, Yonsei University) ;
  • Ji-Woong Yang (Dept. of Artificial Intelligence Semiconductor Engineering, Hanyang University) ;
  • Hyeon-Jin Jung (Dept. of Computer Science, Yonsei University) ;
  • Han-Jin Lee (Dept. of Computer Science, Yonsei University) ;
  • Ellen J. Hong (Division of Software, Yonsei University)
  • Received : 2024.06.11
  • Accepted : 2024.07.26
  • Published : 2024.08.30
Download PDF
⟨ Previous Next ⟩

Abstract

Traffic accidents are not only a threat to human lives but also pose significant societal costs. Recently, research has been conducted to address the issue of traffic accidents by predicting the risk using deep learning technology and spatiotemporal information of roads. However, while traffic accidents are influenced not only by the spatiotemporal information of roads but also by human factors, research on the latter has been relatively less active. This paper analyzes driver groups and characteristics by applying clustering techniques to a traffic accident dataset and proposes and applies a method to calculate the Risk Level for each driver group and characteristic. In this process, the preprocessing technique suggested in this paper demonstrates a higher Silhouette Score of 0.255 compared to the commonly used One-Hot Embedding & Min-Max Scaling techniques, indicating its suitability as a preprocessing method.

교통사고는 인간의 생명뿐만 아니라 사회적으로 큰 비용을 발생시키는 문제이다. 최근에는 교통사고 문제를 해결하기 위하여, 딥러닝 기술과 도로의 시공간적 정보를 통해 교통사고 위험도를 예측하는 연구가 진행되었다. 그러나 교통사고는 도로의 시공간적 정보뿐만 아니라 인적요소 또한 교통사고에 매우 큰 영향을 미치지만 이에 대한 연구는 상대적으로 활성화되지 않았다. 본 논문은 교통사고 데이터셋을 바탕으로 클러스터링 기법을 적용하여 운전자 그룹 및 특성을 분석하였으며, 각 운전자 그룹 및 특성에 대한 위험도를 산출하는 방법을 제시 및 적용하였다. 이 과정에서 본 논문에서 제시한 전처리 기법이 기존에 일반적으로 사용되었던 원-핫 임베딩, Min-Max Scaling 기법보다 더 높은 성능을 보임으로써 더 적합한 전처리 기법임을 보였다.

Keywords

Acknowledgement

This work was supported by a National Research Foundation of Korea (NRF) grant funded by the Korea government (2022R1F1A1074273).

References

  1. "In 2021, about KRW 26,998.7 trillion in social costs due to road traffic accidents was incurred" Accessed: Feb. 13, 2024. [Online]. Available: https://www.koroad.or.kr/main/board/6/87791/board_view.do?&cp=1&listType=list&bdOpenYn=Y&bdNoticeYn=N
  2. "Traffic Accident Analysis System(TAAS) Accessed: Feb. 13, 2024. [Online]. Available: https://taas.koroad.or.kr/web/bdm/srs/selectStaticalReportsList.do?menuId=WEB_KMP_IDA_SRS_TAA
  3. S. Chen, K. Cheng, J. Yang, X. Zang, Q. Luo, and J. Li, "Driving Behavior Risk Measurement and Cluster Analysis Driven by Vehicle Trajectory Data," Applied Sciences, Vol. 13, No. 9, pp. 5675-7495, 2023, doi: 10.3390/APP13095675.
  4. C. Zhang, Y. Ma, A. J. Khattak, S. Chen, G. Xing, and J. Zhang, "Driving style identification and its association with risky driving behaviors among truck drivers based on GPS, load condition, and in-vehicle monitoring data," Journal of Transportation Safety & Security, Vol. 16, No. 5, pp. 507-541, 2024, doi: 10.1080/19439962.2023.2233089
  5. Y. Yao, X. Zhao, Y. Wu, Y. Zhang, and J. Rong, "Clustering driver behavior using dynamic time warping and hidden Markov model," Journal of Intelligent Transportation Systems, Vol. 25, No. 3, pp. 249-262, 2021.
  6. A. R. Anil and J. Anudev, "Driver behavior analysis using K-means algorithm," 3rd International Conference on Intelligent Computing Instrumentation and Control Technologies (ICICICT), pp. 1555-1559, 2022, doi: 10.1109/ICICICT54557.2022.9917899.
  7. Y. Chen, K. Wang, and J. J. Lu, "Feature selection for driving style and skill clustering using naturalistic driving data and driving behavior questionnaire," Accident Analysis & Prevention, vol. 185, pp. 107022-107037, 2023, doi: 10.1016/J.AAP.2023.107022.
  8. J. Lee and K. Jang, "A framework for evaluating aggressive driving behaviors based on in-vehicle driving records," Transporation Research Part F Traffic Psychology Behaviour, vol. 65, pp. 610-619, 2019, doi: 10.1016/J.TRF.2017.11.021.
  9. L. Zhang, H. Wu and K. Cui, "Evaluation of Bus Driving Behavior Based on Vehicle Location Data," 8th International Conference on Image, Vision and Computing (ICIVC), pp. 825-830, 2023, doi:10.1109/ICIVC58118.2023.10270432.
  10. R. Wang, W. Zheng, M. Huang and G. Li, "Driving Behavior Evaluation Based on DBSCAN and Kmeans++ Clustering," 5th International Conference on Advanced Electronic Materials, Computers and Software Engineering (AEMCSE), pp. 188-193, 2022, doi: 10.1109/AEMCSE55572.2022.00046.
  11. K. Yang, C. Al Haddad, G. Yannis and C. Antoniou, "Driving Behavior Safety Levels: Classification and Evaluation," 7th International Conference on Models and Technologies for Intelligent Transportation Systems (MT-ITS), pp. 1-6, 2021, doi:10.1109/MT-ITS49943.2021.9529309.
  12. H. Xiao, Y. Lu, R. Su, B. Wang, N. Zhao and Z. Hu, "Clustering and Analysis of the Driving Style in the Cut-in Process," IEEE 26th International Conference on Intelligent Transportation Systems (ITSC), pp. 3613-3618, 2023, doi: 10.1109/ITSC57777.2023.10421969.
  13. "Traffic Accident Analysis System(TAAS)" Accessed: Feb. 13, 2024. [Online]. Available: https://taas.koroad.orkr/web/shp/sbm/.initGisAnals.do?menuId=WEB_KMP_GIS_TAS
  14. F. Pedregosa FABIANPEDREGOSA et al., "Scikit-learn: Machine Learning in Python," Journal of Machine Learning Research, Vol. 12, pp. 2825-2830, 2011, [Online]. Available:http://jmlr.org/papers/v12/pedregosa11a.html
  15. L. Van Der Maaten and G. Hinton, "Visualizing Data using t-SNE," Journal of Machine Learning Research, Vol. 9, No. 11, pp. 2579-2605, 2008.