گسترش مدل شبکه عصبی گراف پیچشی در مدل‌سازی شدت تصادفات

نوع مقاله : علمی - پژوهشی

نویسندگان
امید عبدالحسین پور محجوبیان 1
علی توکلی کاشانی 2
1 دانش آموخته کارشناسی ارشد، گروه مهندسی حمل و نقل، دانشکده مهندسی عمران، دانشگاه علم و صنعت ایران، ایران
2 استادیار، گروه مهندسی حمل و نقل، دانشکده مهندسی عمران، دانشگاه علم و صنعت ایران، ایران
10.22119/jte.2023.418859.2684
چکیده
به علت خسارات اجتماعی و  اقتصادی وارده توسط تصادفات، مطالعات ایمنی ترافیک در دهه‌های گذشته به‌شدت مورد توجه قرار گرفته‌اند. یکی از ابعاد مطالعات ایمنی ترافیکی، مدل‌سازی شدت تصادفات است که در آن رابطه بین شدت تصادفات و  یک دسته متغیر توضیح مورد بررسی قرار می‌گیرد. علاوه بر مدل‌های آماری مختلفی که به این منظور مورداستفاده قرار گرفته‌اند، مدل‌های یادگیری ماشین نیز در سال‌های اخیر مورد توجه بوده‌اند. شبکه‌های عصبی یکی از این دسته الگوریتم‌های یادگیری ماشین است که در برخی از مطالعات مدل‌سازی شدت تصادفات استفاده شده. اما ازآنجا که روش‌های یادگیری ماشین قابلیت مدل‌سازی مواردی مانند ناهمگونی مشاهده نشده را ندارند و به علت برخی مشکلات دیگر مانند توزیع غیرخطی داده‌ها، استفاده از داده‌های تصادفات به فرم اصلی‌شان همراه با کاستی‌هایی است. در نتیجه پژوهش حاضر با استفاده از الگوریتم نزدیک‌ترین همسایگی، یک ساختار پنهان گراف بر روی داده‌ها ایجاد می‌کند و سپس با استفاده از شبکه عصبی گراف پیچشی به مدل‌سازی شدت تصادفات می‌پردازد. سپس مدل پیشنهادی با چهار مدل یادگیری ماشین از جمله شبکه عصبی پرسپترون چندلایه مقایسه می‌شود نتایج نشان می‌دهد که مدل پیشنهادی در تمام شاخص‌ها از سایر مدل‌ها عملکرد بهتری داشته که می‌توان نتیجه گرفت ساختار پنهان گراف بر روی داده‌های تصادفات توانسته به استخراج روابط معنی‌دارتر از داده‌ها کمک کند.

کلیدواژه‌ها

موضوعات

عنوان مقاله English

Graph Convolutional Neural Networks in Crash Severity Modeling

نویسندگان English

omid abdolhoseinpoor mahjoubian 1
Ali Tavakoli Kashani 2
1 MSc Transportation Engineering, Department of Transportation Engineering, School of Civil Engineering, Iran University of Science and Technology, Tehran, Iran
2 Associate Professor, Department of Transportation Engineering, School of Civil Engineering, Iran University of Science and Technology, Tehran, Iran
چکیده English

Due to high socio-economical costs of traffic crashes, accident injury prediction is an important aspect in safety related studies. Additional to statistical methods that have been traditionally developed in crash severity modeling, some machine learning algorithms have also been employed in recent years. Neural Network being one such algorithm, have gained recognition in the safety related fields in recent field. However, as the distribution of crash data might be nonlinear and due to a number of complications, like inability to model unobserved heterogeneity, using the crash data in its original form for neural networks, may not yield the best result. So, the current study, inspired by the KNN algorithm, develops a hidden graph structure on the data, and then using Graph Convolutional Neural Networks, tries to extract more meaningful relations between crash severity and other explanatory variable. Different metrics adopted for the current task show that the performance of the proposed model is significantly improved over other machine learning models.

کلیدواژه‌ها English

Crash Severity Modeling, Graph Convolutional Network, Hidden Graph Structure
- Abdelwahab, H. T. and M. A. Abdel-Aty (2001). "Development of artificial neural network models to predict driver injury severity in traffic accidents at signalized intersections." Transportation research record 1746(1): 6-13.
 
- Ali, A., Y. Zhu and M. Zakarya (2022). "Exploiting dynamic spatio-temporal graph convolutional neural networks for citywide traffic flows prediction." Neural networks 145: 233-247.
 
- Bai, L., L. Yao, C. Li, X. Wang and C. Wang (2020). "Adaptive graph convolutional recurrent network for traffic forecasting." Advances in neural information processing systems 33: 17804-17815.
 
- Boukerche, A., Y. Tao and P. Sun (2020). "Artificial intelligence-based vehicular traffic flow prediction methods for supporting intelligent transportation systems." Computer networks 182: 107484.
 
- Bui, K.-H. N., J. Cho and H. Yi (2022). "Spatial-temporal graph neural network for traffic forecasting: An overview and open research issues." Applied Intelligence 52(3): 2763-2774.
 
- Chimba, D. and T. Sando (2009). "The prediction of highway traffic accident injury severity with neuromorphic techniques." Advances in transportation studies 19: 17-26.
 
- Delen, D., R. Sharda and M. Bessonov (2006). "Identifying significant predictors of injury severity in traffic accidents using a series of artificial neural networks." Accident Analysis & Prevention 38(3): 434-444.
 
- Do, L. N., H. L. Vu, B. Q. Vo, Z. Liu and D. Phung (2019). "An effective spatial-temporal attention based neural network for traffic flow prediction." Transportation research part C: emerging technologies 108: 12-28.
 
-Fang, S.-H., Y.-X. Fei, Z. Xu and Y. Tsao (2017). "Learning transportation modes from smartphone sensors based on deep neural network." IEEE Sensors Journal 17(18): 6111-6118.
 
- Fanyu, M., N. Sze, S. Cancan, C. Tiantian and Z. Yiping (2021). "Temporal instability of truck volume composition on non-truck-involved crash severity using uncorrelated and correlated grouped random parameters binary logit models with space-time variations." Analytic Methods in Accident Research 31: 100168.
- Haghshenas, S. S., G. Guido, A. Vitale and V. Astarita (2023). "Assessment of the level of road crash severity: Comparison of intelligence studies." Expert Systems with Applications 234: 121118.
 
- Hou, Q., X. Huo, J. Leng and F. Mannering (2022). "A note on out-of-sample prediction, marginal effects computations, and temporal testing with random parameters crash-injury severity models." Analytic methods in accident research 33: 100191.
 
- Hu, Z., J. Zhou, K. Huang and E. Zhang (2022). "A data-driven approach for traffic crash prediction: A case study in Ningbo, China." International Journal of Intelligent Transportation Systems Research 20(2): 508-518.
 
- Iranitalab, A. and A. Khattak (2017). "Comparison of four statistical and machine learning methods for crash severity prediction." Accident Analysis & Prevention 108: 27-36.
 
- Islam, A. M., M. Shirazi and D. Lord (2023). "Grouped Random Parameters Negative Binomial-Lindley for accounting unobserved heterogeneity in crash data with preponderant zero observations." Analytic methods in accident research 37: 100255.
 
- Islam, M. R., K. B. Kelarestaghi, A. Ermagun and S. Banerjee (2019). "Gender Differences in Injury Severity Risk of Single-Vehicle Crashes in Virginia: A Nested Logit Analysis of Heterogeneity." arXiv preprint arXiv:1901.03289.
 
- Li, L., J. Zhu, H. Zhang, H. Tan, B. Du and B. Ran (2020). "Coupled application of generative -- adversarial networks and conventional neural networks for travel mode detection using GPS data." Transportation Research Part A: Policy and Practice 136: 282-292.
- Li, P., M. Abdel-Aty and J. Yuan (2020). "Real-time crash risk prediction on arterials based on LSTM-CNN." Accident Analysis & Prevention 135: 105371.
 
- Organization, W. H. (2020). Road safety in the Eastern Mediterranean Region: facts from the global status report on road safety 2018, World Health Organization. Regional Office for the Eastern Mediterranean.
 
- Park, S. and J. Park (2022). "Multilevel Mixed-Effects Models to Identify Contributing Factors on Freight Vehicle Crash Severity." Sustainability 14(19): 11804.
 
- Patil, S., S. R. Geedipally and D. Lord (2012). "Analysis of crash severities using nested logit model—accounting for the underreporting of crashes." Accident Analysis & Prevention 45: 646-653.
 
- Rahim, M. A. and H. M. Hassan (2021). "A deep learning based traffic crash severity prediction framework." Accident Analysis & Prevention 154: 106090.
 
- Rezapour, M., S. Nazneen and K. Ksaibati (2020). "Application of deep learning techniques in predicting motorcycle crash severity." Engineering Reports 2(7): e12175.
 
- Rezapour, M., S. S. Wulff and K. Ksaibati (2020). "Bayesian hierarchical modelling of traffic barrier crash severity." International journal of injury control and safety promotion 28(1): 94-102.
 
- Sameen, M. I. and B. Pradhan (2017). "Severity prediction of traffic accidents with recurrent neural networks." Applied Sciences 7(6): 476.
 
- Silva, P. B., M. Andrade and S. Ferreira (2020). "Machine learning applied to road safety modeling: A systematic literature review." Journal of traffic and transportation engineering (English edition) 7(6): 775-790.
 
- Vajari, M. A., K. Aghabayk, M. Sadeghian and N. Shiwakoti (2020). "A multinomial logit model of motorcycle crash severity at Australian intersections." Journal of safety research 73: 17-24.
 
- Yu, H., Z. Li, G. Zhang, P. Liu and T. Ma (2021). "Fusion convolutional neural network-based interpretation of unobserved heterogeneous factors in driver injury severity outcomes in single-vehicle crashes." Analytic methods in accident research 30: 100157.
 
- Yu, L., B. Du, X. Hu, L. Sun, L. Han and W. Lv (2021). "Deep spatio-temporal graph convolutional network for traffic accident prediction." Neurocomputing 423: 135-147.
 
- Zeng, Q., W. Gu, X. Zhang, H. Wen, J. Lee and W. Hao (2019). "Analyzing freeway crash severity using a Bayesian spatial generalized ordered logit model with conditional autoregressive priors." Accident Analysis & Prevention 127: 87-95.
 
- Zeng, Q. and H. Huang (2014). "A stable and optimized neural network model for crash injury severity prediction." Accident Analysis & Prevention 73: 351-358.
 
- Zeng, Q., H. Wen, H. Huang, X. Pei and S. Wong (2017). "A multivariate random-parameters Tobit model for analyzing highway crash rates by injury severity." Accident Analysis & Prevention 99: 184-191.
 
- Zhang, S., H. Tong, J. Xu and R. Maciejewski (2019). "Graph convolutional networks: a comprehensive review." Computational Social Networks 6(1): 1-23.
 
-Zhang, Y., X. Dong, L. Shang, D. Zhang and D. Wang (2020). A multi-modal graph neural network approach to traffic risk forecasting in smart urban sensing. 2020 17th Annual IEEE international conference on sensing, communication, and networking (SECON), IEEE.
 
- Zhou, F., Q. Yang, K. Zhang, G. Trajcevski, T. Zhong and A. Khokhar (2020). "Reinforced spatiotemporal attentive graph neural networks for traffic forecasting." IEEE Internet of Things Journal 7(7): 6414-6428.
 
- Zhou, F., Q. Yang, T. Zhong, D. Chen and N. Zhang (2020). "Variational graph neural networks for road traffic prediction in intelligent transportation systems." IEEE Transactions on Industrial Informatics 17(4): 2802-2812.
 
- Zhou, Z., Y. Wang, X. Xie, L. Chen and H. Liu (2020). RiskOracle: A minute-level citywide traffic accident forecasting framework. Proceedings of the AAAI conference on artificial intelligence.
 
- Zhou, Z., Y. Wang, X. Xie, L. Chen and C. Zhu (2020). "Foresee urban sparse traffic accidents: A spatiotemporal multi-granularity perspective." IEEE Transactions on Knowledge and Data Engineering 34(8): 3786-3799.
 
- Zubaidi, H. A., I. A. Obaid, A. Alnedawi and S. Das (2021). "Motor vehicle driver injury severity analysis utilizing a random parameter binary probit model considering different types of driving licenses in 4-legs roundabouts in South Australia." Safety science 134: 105083.
 
- Zuo, D., C. Qian, D. Xiao, X. Xu and H. Wang (2023). "Data-driven crash prediction by injury severity using a recurrent neural network model based on Keras framework." International Journal of Injury Control and Safety Promotion: 1-10.

  • تاریخ دریافت 09 مهر 1402
  • تاریخ بازنگری 17 مهر 1402
  • تاریخ پذیرش 18 مهر 1402