Traffic and Transportation Journal

Article

Analysis of Beijing Traffic Violations Based on the BERT-CRF Model

Jie Li, Yuntao Shi, Shuqin Li

DOI:https://doi.org/10.7307/ptt.v36i2.366

Keywords:traffic violation, traffic accident, name entity, People's Daily, BERT-CRF, Bayesian formula

Abstract

Traffic violations are a major cause of traffic accidents, yet current research falls short in comprehensively analysing these violations and the named entity method fails to extract the name of traffic violation events from records, thereby lacking in providing guidance for managing urban traffic violations. By expanding the People’s Daily dataset from 71,456 words to 95,291 words, the BERT-CRF (Bidirectional Encoder Representations from Transformers-Conditional Random Field) model achieves an accuracy rate of 88.53%, a recall rate of 92.90% and an F1 score of 90.66%, successfully identifying event, time and location named entities within traffic violations. The data of traffic violations is then enhanced through forward geocoding and the Bayesian formula, and traffic violations are analysed from time, space, administrative region, gender and weather, to provide support for the dynamic allocation of law enforcement forces on traffic scenes and the precise management of
traffic violations.

References

[1] National Bureau of Statistics. China statistical yearbook. Beijing: China Statistics Press;2021. https://www.stats.gov.cn/sj/ndsj/2021/indexch.htm [Accessed 10th Sep. 2021].
[2] Yigitcanlar T, Li R, Inkinen T, Paz A. Public perceptions on application areas and adoption challenges of AI in urban services. Emerging Science Journal. 2022. DOI: 10.28991/ESJ-2022-06-06-01.
[3] Jamshid A, Kudratulla A, Ilkhomjon S. Method of analysis of the reasons and consequences of traffic accidents in Uzbekistan cities. International Journal of Safety and Security Engineering. 2020;10(4):483-490. DOI: 10.18280/ijsse.100407.
[4] Sorum NG, Pal D. Effect of distracting factors on driving performance: A review. Civil Engineering Journal. 2022. DOI: 10.28991/cej-2022-08-02-014.
[5] Abbas HA, Obaid HA, Alwash A. Enhanced road network to reduce the effect of (external – external) freight trips on traffic flow. Civil Engineering Journal. 2022. DOI: 10.28991/cej-2022-08-11-015.
[6] Wang C, Hu HT, Deng SH. Development of a knowledge base for reasoning penalty for traffic violations based on event evolutionary graph. Journal of Transport Information and Safety. 2022;40(1):36-44. DOI: 10.3963/j.jssn.1674-4861.2022.01.005.
[7] Zhao ZY, et al. Study on the method of identifying the characteristics of the traffic violation behavior based on the spatial and temporal hotspot analysis approach. Journal of Geo-Information Science. 2022;24(7):1312-1325. DOI: 10.12082/dqxxkx.2022.210599.
[8] Li YX, et al. Analysis and prediction of intersection traffic violations using automated enforcement system data. Accident Analysis & Prevention. 2021;162:106422. DOI: 10.1016/j.aap.2021.106422.
[9] Ji ZY, et al. Named entity recognition based on deep learning. Computer Integrated Manufacturing Systems. 2022;28(6):1603-1615. DOI: 10.13196/j.cims.2022.06.001.
[10] Xiao R, Hu FJ, Pei W. Chinese medicine text named entity recognition based on BiLSTM-CRF. World Science and Technology-Modernization of Traditional Chinese Medicine. 2020;22(07):2504-2510. DOI: 10.1155/2021/6696205.
[11] Chen ZL, Yuan F, Li XH, Zhang MM. Based on BERT-BiLSTM-CRF model the named entity and relation joint extraction of Chinese lithological description corpus. Geological Review. 2022;68(02):742-750. DOI: 10.16509/j.georeview.2022.01.115.
[12] Zhao PF, Zhao CJ, Wu HR, Wang W. Recognition of the agricultural named entities with multi-feature fusion based on BERT. Transactions of the Chinese Society of Agricultural Engineering. 2022;38(3):112-118. DOI: 10.11975/j.issn.1002-6819.2022.03.013.
[13] Zeng LL, Wang YS, Chen PF. Named entity recognition based on BERT and joint learning for judgment documents. Journal of Computer Applications. 2022;1-7. DOI: 10.11772/j.issn.1001-9081.2021091565.
[14] Liu F, Wen Z, Wu Y. Intelligent analysis on text of power industry accident based on BERT-BILSTM-CRF model. Journal of Safety Science and Technology. 2023;19(01):209-215. DOI: 10.11731/j.issn.1673-193x.2023.01.031.
[15] Wang ZH, et al. Intelligent recognition of key earthquake emergency Chinese information based on the optimized BERT-BiLSTM-CRF algorithm. Applied Sciences. 2023;13(5):3024. DOI: 10.3390/app13053024.
[16] Li SQ, Pang WT. Joint extraction method of entity and relation in maize breeding based on BERT-CRF and word embedding. Transactions of the Chinese Society for Agricultural Machinery. 2023;1-16. DOI: 10.6041/j.issn.1000-1298.2023.11.028.
[17] Li J, Shi YT, Li SQ, Wang Q. Construction of knowledge graph based on traffic violations in Beijing. 2022 4th International Conference on Intelligent Information Processing (IIP), 14-16 Oct. 2022, Guangzhou, China. 2022. p. 113-116. DOI: 10.1109/IIP57348.2022.00030.
[18] Devlin J, Chang MW, Lee K, Toutanova K. BERT: Pre-training deep bidirectional transformers for language understanding. Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, June 2019, Minneapolis, Minnesota. 2019. p. 4171-4186. DOI: 10.18653/v1/N19-1423.
[19] Lafferty JD, McCallum A, Pereira F. Conditional random fields: Probabilistic models for segmenting and labeling sequence data. Proceedings of the Eighteenth International Conference on Machine Learning (ICML 2001), Williams College, Williamstown, MA, USA, June 28 - July 1, 2001. 2001. p. 282-289. DOI: 10.1109/ICIP.2012.6466940.
[20] Vaswani A, et al. Attention is all you need. Proceedings of the 31st International Conference on Neural Information Processing Systems (NIPS'17), Red Hook, NY, USA. 2017. p. 6000-6010. DOI: 10.48550/arXiv.1706.03762.
[21] Rolison J, Regev S, Moutari S, Feeney A. What are the factors that contribute to road accidents? An assessment of law enforcement views, ordinary drivers' opinions, and road accident records. Accident Analysis & Prevention. 2018;115:11-24. DOI: 10.1016/j.aap.2018.02.025.
[22] Regev S, Rolison J, Moutari S. Crash risk by driver age, gender, and time of day using a new exposure methodology. Journal of Safety Research. 2018;66:131-140. DOI: 10.1016/j.jsr.2018.07.002.
[23] Wu YN, Abdel-Aty M, Lee J. Crash risk analysis during fog conditions using real-time traffic data. Accident Analysis & Prevention. 2018;114:4-11. DOI: 10.1016/j.aap.2017.05.004.
[24] Theofilatos A, Yannis G. A review of the effect of traffic and weather characteristics on road safety. Accident Analysis & Prevention. 2014;72:244-256. DOI: 10.1016/j.aap.2014.06.017.
[25] Xing F, et al. Hourly associations between weather factors and traffic crashes: Non-linear and lag effects. Analytic Methods in Accident Research. 2019;24:100-109. DOI: 10.1016/j.amar.2019.100109.
[26] Oppenheim I, Oron-Gilad T, Parmet Y, Shinar D. Can traffic violations be traced to gender-role, sensation seeking, demographics and driving exposure?. Transportation Research Part F: Traffic Psychology and Behaviour. 2016;43:387-395. DOI: 10.1016/J.TRF.2016.06.027.
[27] Ozkan T, et al. Cross-cultural differences in driving skills: A comparison of six countries. Accident Analysis & Prevention. 2006;38(5):1011-1018. DOI: 10.1016/j.aap.2006.04.006.
[28] Li XM, Oviedo-Trespalacios O, Rakotonirainy A, Yan XD. Collision risk management of cognitively distracted drivers in a car-following situation. Transportation Research Part F: Traffic Psychology and Behaviour. 2019;60:288-298. DOI: 10.1016/J.TRF.2018.10.011.

PDF

Published

30.04.2024

Issue

2024 (Vol 36), Issue 2

This work is licensed under a
Creative Commons Attribution-NonCommercial-
NoDerivatives 4.0 International License.