{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,15]],"date-time":"2026-09-15T17:04:20Z","timestamp":1789491860604,"version":"build-2803163510"},"reference-count":38,"publisher":"Elsevier BV","license":[{"start":{"date-parts":[[2026,9,1]],"date-time":"2026-09-01T00:00:00Z","timestamp":1788220800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/tdm\/userlicense\/1.0\/"},{"start":{"date-parts":[[2026,9,1]],"date-time":"2026-09-01T00:00:00Z","timestamp":1788220800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/legal\/tdmrep-license"}],"funder":[{"DOI":"10.13039\/100014793","name":"Kentucky Transportation Cabinet","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100014793","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["elsevier.com","sciencedirect.com"],"crossmark-restriction":true},"short-container-title":["Journal of Safety Research"],"published-print":{"date-parts":[[2026,9]]},"DOI":"10.1016\/j.jsr.2026.08.017","type":"journal-article","created":{"date-parts":[[2026,8,27]],"date-time":"2026-08-27T06:50:40Z","timestamp":1787813440000},"page":"588-598","update-policy":"https:\/\/doi.org\/10.1016\/elsevier_cm_policy","source":"Crossref","is-referenced-by-count":0,"special_numbering":"C","title":["Improving crash data quality with large language models: Evidence from secondary crash narratives in Kentucky"],"prefix":"10.1016","volume":"98","author":[{"given":"Xu","family":"Zhang","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mei","family":"Chen","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"78","reference":[{"key":"10.1016\/j.jsr.2026.08.017_b0005","doi-asserted-by":"crossref","first-page":"3","DOI":"10.1016\/j.aap.2017.07.018","article-title":"Errors in accident data, its types, causes and methods of rectification-analysis of the literature","volume":"130","author":"Ahmed","year":"2019","journal-title":"Accident Analysis & Prevention"},{"key":"10.1016\/j.jsr.2026.08.017_b0010","series-title":"Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining","article-title":"Optuna: A next-generation hyperparameter optimization framework","author":"Akiba","year":"2019"},{"key":"10.1016\/j.jsr.2026.08.017_b0015","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1016\/j.jsr.2024.11.009","article-title":"A large language model framework to uncover underreporting in traffic crashes","volume":"92","author":"Arteaga","year":"2025","journal-title":"Journal of Safety Research"},{"key":"10.1016\/j.jsr.2026.08.017_b0020","unstructured":"Beltagy, I., M.E. Peters, and A. Cohan, (2020). Longformer: The long-document transformer. arXiv preprint arXiv:2004.05150."},{"key":"10.1016\/j.jsr.2026.08.017_b0025","doi-asserted-by":"crossref","unstructured":"Bhagat, S.R., I.F. Shihab, and A. Sharma, (2025). Accuracy is Not Agreement: Expert-Aligned Evaluation of Crash Narrative Classification Models. arXiv preprint arXiv:2504.13068.","DOI":"10.1109\/ITSC60802.2025.11423032"},{"key":"10.1016\/j.jsr.2026.08.017_b0030","series-title":"in Proceedings of the 2019 conference of the North American chapter of the association for computational linguistics: human language technologies, volume 1 (long and short papers)","article-title":"Pre-training of deep bidirectional transformers for language understanding","author":"Devlin","year":"2019"},{"key":"10.1016\/j.jsr.2026.08.017_b0035","unstructured":"Dubey, A., et al., (2024). The llama 3 herd of models. arXiv e-prints, 2024: p. arXiv: 2407.21783."},{"key":"10.1016\/j.jsr.2026.08.017_b0040","unstructured":"Fields, M.A., et al., (2024). Pilot Study on Improving Crash Data Accuracy in Kentucky through University Collaboration, Kentucky Transportation Center. p. 30p."},{"key":"10.1016\/j.jsr.2026.08.017_b0045","doi-asserted-by":"crossref","first-page":"122","DOI":"10.1016\/j.aap.2017.08.026","article-title":"Construction accident narrative classification: An evaluation of text mining techniques","volume":"108","author":"Goh","year":"2017","journal-title":"Accident Analysis & Prevention"},{"key":"10.1016\/j.jsr.2026.08.017_b0050","unstructured":"Guo D, et al., (2025). Deepseek-r1: Incentivizing reasoning capability in llms via reinforcement learning. arXiv preprint arXiv:2501.12948."},{"key":"10.1016\/j.jsr.2026.08.017_b0055","doi-asserted-by":"crossref","unstructured":"Heidarysafa, M., et al. (2018). Analysis of Railway Accidents' Narratives Using Deep Learning. in 2018 17th IEEE International Conference on Machine Learning and Applications (ICMLA). 2018. IEEE.","DOI":"10.1109\/ICMLA.2018.00235"},{"issue":"4","key":"10.1016\/j.jsr.2026.08.017_b0060","doi-asserted-by":"crossref","first-page":"1038","DOI":"10.1016\/j.ijtst.2022.12.002","article-title":"Application of text mining techniques to identify actual wrong-way driving (WWD) crashes in police reports","volume":"12","author":"Hosseini","year":"2023","journal-title":"International Journal of Transportation Science and Technology"},{"key":"10.1016\/j.jsr.2026.08.017_b0065","unstructured":"HuggingFace (2025). Transformers. [cited 2025 7\/31\/2025]; Available from: https:\/\/huggingface.co\/docs\/transformers\/en\/index."},{"key":"10.1016\/j.jsr.2026.08.017_b0070","doi-asserted-by":"crossref","first-page":"84","DOI":"10.1016\/j.aap.2017.02.022","article-title":"Crash data quality for road safety research: Current state and future directions","volume":"130","author":"Imprialou","year":"2019","journal-title":"Accident Analysis & Prevention"},{"key":"10.1016\/j.jsr.2026.08.017_b0075","doi-asserted-by":"crossref","unstructured":"Karim, M.M., et al., (2025). Large Language Models and Their Applications in Roadway Safety and Mobility Enhancement: A Comprehensive Review. arXiv preprint arXiv:2506.06301.","DOI":"10.1016\/j.ait.2025.100004"},{"issue":"1","key":"10.1016\/j.jsr.2026.08.017_b0080","doi-asserted-by":"crossref","first-page":"74","DOI":"10.1080\/15389588.2020.1836365","article-title":"Crash narrative classification: Identifying agricultural crashes using machine learning with curated keywords","volume":"22","author":"Kim","year":"2021","journal-title":"Traffic Injury Prevention"},{"key":"10.1016\/j.jsr.2026.08.017_b0085","unstructured":"Liu Y, et al., (2019). Roberta: A robustly optimized bert pretraining approach. arXiv preprint arXiv:1907.11692."},{"key":"10.1016\/j.jsr.2026.08.017_b0090","unstructured":"Mikolov, T., et al., (2013). Efficient estimation of word representations in vector space. arXiv preprint arXiv:1301.3781."},{"key":"10.1016\/j.jsr.2026.08.017_b0095","unstructured":"Mumtarin, M., M.S. Chowdhury, and J. Wood, (2023). Large Language Models in Analyzing Crash Narratives--A Comparative Study of ChatGPT, BARD and GPT-4. arXiv preprint arXiv:2308.13563."},{"key":"10.1016\/j.jsr.2026.08.017_b0100","doi-asserted-by":"crossref","DOI":"10.1016\/j.nlp.2023.100007","article-title":"Using Bidirectional Encoder Representations from Transformers (BERT) to classify traffic crash severity types","volume":"3","author":"Oliaee","year":"2023","journal-title":"Natural Language Processing Journal"},{"issue":"2","key":"10.1016\/j.jsr.2026.08.017_b0105","doi-asserted-by":"crossref","first-page":"83","DOI":"10.1177\/03611981241260691","article-title":"Automating pedestrian crash typology using transformer models","volume":"2679","author":"Oliaee","year":"2025","journal-title":"Transport Research Record"},{"key":"10.1016\/j.jsr.2026.08.017_b0110","series-title":"in Proceedings of the 2014 conference on empirical methods in natural language processing (EMNLP)","article-title":"Glove: Global vectors for word representation","author":"Pennington","year":"2014"},{"key":"10.1016\/j.jsr.2026.08.017_b0115","doi-asserted-by":"crossref","first-page":"11","DOI":"10.1016\/j.aap.2018.02.025","article-title":"What are the factors that contribute to road accidents? an assessment of law enforcement views, ordinary drivers\u2019 opinions, and road accident records","volume":"115","author":"Rolison","year":"2018","journal-title":"Accident Analysis & Prevention"},{"key":"10.1016\/j.jsr.2026.08.017_b0120","unstructured":"Sanh V, et al., (2019). DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter. arXiv preprint arXiv:1910.01108."},{"key":"10.1016\/j.jsr.2026.08.017_b0125","doi-asserted-by":"crossref","DOI":"10.1016\/j.aap.2021.106211","article-title":"Identification and analysis of misclassified work-zone crashes using text mining techniques","volume":"159","author":"Sayed","year":"2021","journal-title":"Accident Analysis & Prevention"},{"key":"10.1016\/j.jsr.2026.08.017_b0130","unstructured":"scikit-learn.org, (2025). Logistic Regression. [cited 2025 7\/31\/2025]; Available from: https:\/\/scikit-learn.org\/stable\/modules\/generated\/sklearn.linear_model.LogisticRegression.html."},{"key":"10.1016\/j.jsr.2026.08.017_b0135","doi-asserted-by":"crossref","DOI":"10.1016\/j.aap.2023.107426","article-title":"Police and hospital data linkage for traffic injury surveillance: A systematic review","volume":"197","author":"Soltani","year":"2024","journal-title":"Accident Analysis & Prevention"},{"issue":"1","key":"10.1016\/j.jsr.2026.08.017_b0140","doi-asserted-by":"crossref","first-page":"131","DOI":"10.1016\/0001-4575(95)00055-0","article-title":"Motor vehicle crashes in roadway construction workzones: An analysis using narrative text from insurance claims","volume":"28","author":"Sorock","year":"1996","journal-title":"Accident Analysis & Prevention"},{"key":"10.1016\/j.jsr.2026.08.017_b0145","doi-asserted-by":"crossref","first-page":"80","DOI":"10.1016\/j.compind.2015.09.005","article-title":"Natural language processing for aviation safety reports: From classification to interactive analysis","volume":"78","author":"Tanguy","year":"2016","journal-title":"Computers in Industry"},{"key":"10.1016\/j.jsr.2026.08.017_b0150","unstructured":"Team G, et al., (2025). Gemma 3 technical report. arXiv preprint arXiv:2503.19786."},{"key":"10.1016\/j.jsr.2026.08.017_b0155","unstructured":"Vaswani, A., et al., (2017). Attention is all you need, in Proceedings of the 31st International Conference on Neural Information Processing Systems. 2017, Curran Associates Inc.: Long Beach, California, USA. p. 6000\u20136010."},{"key":"10.1016\/j.jsr.2026.08.017_b0160","doi-asserted-by":"crossref","first-page":"40","DOI":"10.1016\/j.aap.2018.04.001","article-title":"Methodological evolution and frontiers of identifying, modeling and preventing secondary crashes on highways","volume":"117","author":"Yang","year":"2018","journal-title":"Accident Analysis & Prevention"},{"key":"10.1016\/j.jsr.2026.08.017_b0165","article-title":"Xlnet: Generalized autoregressive pretraining for language understanding","volume":"32","author":"Yang","year":"2019","journal-title":"Advances in Neural Information Processing Systems"},{"key":"10.1016\/j.jsr.2026.08.017_b0170","unstructured":"Yang A, et al., (2025). Qwen3 technical report. arXiv preprint arXiv:2505.09388."},{"issue":"10","key":"10.1016\/j.jsr.2026.08.017_b0175","doi-asserted-by":"crossref","first-page":"1338","DOI":"10.1080\/19439962.2019.1597795","article-title":"Identifying secondary crashes using text mining techniques","volume":"12","author":"Zhang","year":"2020","journal-title":"Journal of Transportation Safety & Security"},{"issue":"9","key":"10.1016\/j.jsr.2026.08.017_b0180","doi-asserted-by":"crossref","first-page":"167","DOI":"10.1177\/03611981211001077","article-title":"Collection, analysis, and reporting of Kentucky traffic incident management performance","volume":"2675","author":"Zhang","year":"2021","journal-title":"Transportation Research Record: Journal of the Transportation Research Board"},{"key":"10.1016\/j.jsr.2026.08.017_b0185","doi-asserted-by":"crossref","first-page":"117","DOI":"10.3141\/2514-13","article-title":"Analyses of multiyear statewide secondary crash data and automatic crash report reviewing","volume":"2514","author":"Zheng","year":"2015","journal-title":"Transportation Research Record: Journal of the Transportation Research Board"},{"issue":"4","key":"10.1016\/j.jsr.2026.08.017_b0190","doi-asserted-by":"crossref","DOI":"10.1093\/tse\/tdae018","article-title":"Crash-causing information extraction via text-mining techniques: Implementation of the chinese state-related crash narratives","volume":"6","author":"Zou","year":"2024","journal-title":"Transportation Safety and Environment"}],"container-title":["Journal of Safety Research"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S0022437526001568?httpAccept=text\/xml","content-type":"text\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S0022437526001568?httpAccept=text\/plain","content-type":"text\/plain","content-version":"vor","intended-application":"text-mining"}],"deposited":{"date-parts":[[2026,9,15]],"date-time":"2026-09-15T16:25:14Z","timestamp":1789489514000},"score":1,"resource":{"primary":{"URL":"https:\/\/linkinghub.elsevier.com\/retrieve\/pii\/S0022437526001568"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9]]},"references-count":38,"alternative-id":["S0022437526001568"],"URL":"https:\/\/doi.org\/10.1016\/j.jsr.2026.08.017","relation":{},"ISSN":["0022-4375"],"issn-type":[{"value":"0022-4375","type":"print"}],"subject":[],"published":{"date-parts":[[2026,9]]},"assertion":[{"value":"Elsevier","name":"publisher","label":"This article is maintained by"},{"value":"Improving crash data quality with large language models: Evidence from secondary crash narratives in Kentucky","name":"articletitle","label":"Article Title"},{"value":"Journal of Safety Research","name":"journaltitle","label":"Journal Title"},{"value":"https:\/\/doi.org\/10.1016\/j.jsr.2026.08.017","name":"articlelink","label":"CrossRef DOI link to publisher maintained version"},{"value":"article","name":"content_type","label":"Content Type"},{"value":"\u00a9 2026 National Safety Council and Elsevier Ltd. All rights are reserved, including those for text and data mining, AI training, and similar technologies.","name":"copyright","label":"Copyright"}]}}