{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,29]],"date-time":"2026-03-29T11:00:51Z","timestamp":1774782051061,"version":"3.50.1"},"reference-count":47,"publisher":"MDPI AG","issue":"6","license":[{"start":{"date-parts":[[2025,5,27]],"date-time":"2025-05-27T00:00:00Z","timestamp":1748304000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Algorithms"],"abstract":"<jats:p>Ensuring high data quality in traffic crash datasets is critical for effective safety analysis and policymaking. This study presents an AI-assisted framework for auditing crash data integrity by detecting potentially mislabeled records related to construction zone (czone) involvement. A GPT-3.5 model was fine-tuned using a fusion of structured crash attributes and unstructured narrative text (i.e., multimodal input) to predict work zone involvement. The model was applied to 6400 crash reports to flag discrepancies between predicted and recorded labels. Among 80 flagged mismatches, expert review confirmed four records as genuine misclassifications, demonstrating the framework\u2019s capacity to surface high-confidence labeling errors. The model achieved strong overall accuracy (98.75%) and precision (86.67%) for the minority class, but showed low recall (14.29%), reflecting its conservative design that minimizes false positives in an imbalanced dataset. This precision-focused approach supports its use as a semi-automated auditing tool, capable of narrowing the scope for expert review and improving the reliability of large-scale traffic safety datasets. The framework is also adaptable to other misclassified crash attributes or domains where structured and unstructured data can be fused for data quality assurance.<\/jats:p>","DOI":"10.3390\/a18060317","type":"journal-article","created":{"date-parts":[[2025,5,27]],"date-time":"2025-05-27T05:52:52Z","timestamp":1748325172000},"page":"317","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":8,"title":["AI for Data Quality Auditing: Detecting Mislabeled Work Zone Crashes Using Large Language Models"],"prefix":"10.3390","volume":"18","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-6322-2792","authenticated-orcid":false,"given":"Shadi","family":"Jaradat","sequence":"first","affiliation":[{"name":"Australian International Institute of Higher Education, Brisbane 4000, Australia"},{"name":"Centre of Data Science, Queensland University of Technology, Brisbane 4000, Australia"},{"name":"Cogninet Australia, Sydney 2010, Australia"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2132-5184","authenticated-orcid":false,"given":"Nirmal","family":"Acharya","sequence":"additional","affiliation":[{"name":"Australian International Institute of Higher Education, Brisbane 4000, Australia"},{"name":"Cogninet Australia, Sydney 2010, Australia"},{"name":"School of Business and Law, Central Queensland University, Brisbane 4000, Australia"}]},{"given":"Smitha","family":"Shivshankar","sequence":"additional","affiliation":[{"name":"Australian International Institute of Higher Education, Brisbane 4000, Australia"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6388-0904","authenticated-orcid":false,"given":"Taqwa I.","family":"Alhadidi","sequence":"additional","affiliation":[{"name":"Faculty of Engineering, Hourani Center for Applied Scientific Research, Al-Ahliyya Amman University, Amman 19111, Jordan"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2634-4576","authenticated-orcid":false,"given":"Mohammad","family":"Elhenawy","sequence":"additional","affiliation":[{"name":"Centre of Data Science, Queensland University of Technology, Brisbane 4000, Australia"}]}],"member":"1968","published-online":{"date-parts":[[2025,5,27]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"377","DOI":"10.1177\/03611981221092716","article-title":"Work Zone Crash Occurrence Prediction Based on Planning Stage Work Zone Configurations Using an Artificial Neural Network","volume":"2676","author":"Cheng","year":"2022","journal-title":"Transp. Res. Rec."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"387","DOI":"10.1080\/15389588.2014.948615","article-title":"Work Zone Safety Analysis and Modeling: A State-of-the-Art Review","volume":"16","author":"Yang","year":"2015","journal-title":"Traffic Inj. Prev."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"222","DOI":"10.1080\/15389588.2020.1734190","article-title":"Understanding Vehicle Crashes in Work Zones: Analysis of Workplace Health and Safety Data as an Alternative to Police-Reported Crash Data in Queensland, Australia","volume":"21","author":"Blackman","year":"2020","journal-title":"Aust. Traffic Inj. Prev."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"106211","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":"Accid. Anal. Prev."},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Almahdi, A., Al Mamlook, R.E., Bandara, N., Almuflih, A.S., Nasayreh, A., Gharaibeh, H., Alasim, F., Aljohani, A., and Jamal, A. (2023). Boosting Ensemble Learning for Freeway Crash Classification under Varying Traffic Conditions: A Hyperparameter Optimization Approach. Sustainability, 15.","DOI":"10.3390\/su152215896"},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"60","DOI":"10.3141\/2237-07","article-title":"Estimation of Real-Time Crash Risk","volume":"2237","author":"Pande","year":"2011","journal-title":"Transp. Res. Rec."},{"key":"ref_7","unstructured":"OpenAI (2023). GPT-3.5 Turbo Fine-Tuning and API Updates, OpenAI."},{"key":"ref_8","unstructured":"Swansen, E., Mckinnon, I.A., and Knodler, M.A. (2013, January 13\u201317). Integration of Crash Report Narratives for Identification of Work Zone-Related Crash Classification. Proceedings of the Transportation Research Board 92nd Annual Meeting, Washington, DC, USA."},{"key":"ref_9","unstructured":"Carrick, G., Heaslip, K., Srinivasan, S., and Brady, B. (2009, January 11\u201315). A Case Study in Spatial Misclassification of Work Zone Crashes. Proceedings of the 88th Transportation Research Board Annual Meeting, National Academy of Sciences, Washington, DC, USA."},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Asadi, H., and Wang, J. (2023). An Ensemble Approach for Predicting Crash Severity in Work Zones Using Machine Learning. Sustainability, 15.","DOI":"10.3390\/su15119076"},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"e13833","DOI":"10.1111\/exsy.13833","article-title":"A Systematic Re-view on Text Summarization: Techniques, Challenges, Opportunities","volume":"42","author":"Sharma","year":"2025","journal-title":"Expert Syst."},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Nusir, M., Louati, A., Louati, H., Tariq, U., Zitar, R.A., Abualigah, L., and Gandomi, A.H. (2022). Design Research Insights on Text Mining Analysis: Establishing the Most Used and Trends in Keywords of Design Research Journals. Electronics, 11.","DOI":"10.3390\/electronics11233930"},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Jaradat, S., Elhenawy, M., Nayak, R., Paz, A., Ashqar, H.I., and Glaser, S. (2025). Multimodal Data Fusion for Tabular and Textual Data: Zero-Shot, Few-Shot, and Fine-Tuning of Generative Pre-Trained Transformer Models. AI, 6.","DOI":"10.3390\/ai6040072"},{"key":"ref_14","unstructured":"Alhadidi, T.I., Alazmi, A., Jaradat, S., Jaber, A., Ashqar, H., and Elhenawy, M. (arXiv, 2025). Pavement Distress Classification Using Bidirectional Cascaded Neural Networks (BCNNs) and U-Net 50-Based Augmented Datasets, arXiv, in press."},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Chang, Y., and Edara, P. (2017, January 16\u201319). Predicting Hazardous Events in Work Zones Using Naturalistic Driving Data. Proceedings of the 2017 IEEE 20th International Conference on Intelligent Transportation Systems (ITSC), Yokohama, Japan.","DOI":"10.1109\/ITSC.2017.8317847"},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"107361","DOI":"10.1016\/j.aap.2023.107361","article-title":"Toward Safer Highway Work Zones: An Empirical Analysis of Crash Risks Using Improved Safety Potential Field and Machine Learning Techniques","volume":"194","author":"Wang","year":"2024","journal-title":"Accid. Anal. Prev."},{"key":"ref_17","first-page":"768","article-title":"Exploring Contributing Factors to Improper Driving Actions in Single-Vehicle Work Zone Crashes: A Mixed Logit Analysis Considering Heterogeneity in Means and Variances, and Temporal Instability","volume":"16","author":"Rangaswamy","year":"2023","journal-title":"J. Transp. Saf. Secur."},{"key":"ref_18","unstructured":"Mashhadi, A.H., Rashidi, A., Medina, J., and Markovi\u0107, N. (2023). Comparing Performance of Different Machine Learning Methods for Predicting Severity of Construction Work Zone Crashes. Comput. Civ. Eng."},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"180","DOI":"10.3141\/1897-23","article-title":"Revisiting the Influence of Crash Report Forms on Work Zone Crash Data","volume":"1897","author":"Ullman","year":"2004","journal-title":"Transp. Res. Rec."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"61","DOI":"10.3141\/2485-08","article-title":"Exploration of Work Zone Crash Causes and Implications for Safety Performance Measurement Programs","volume":"2485","author":"Clark","year":"2015","journal-title":"Transp. Res. Rec."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"18","DOI":"10.3141\/1715-03","article-title":"Analysis of Fatal Crashes in Georgia Work Zones","volume":"1715","author":"Daniel","year":"2000","journal-title":"Transp. Res. Rec."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"54","DOI":"10.1177\/0361198196152900107","article-title":"Investigation of Highway Work Zone Crashes: What We Know and What We Don\u2019t Know","volume":"1529","author":"Wang","year":"1996","journal-title":"Transp. Res. Rec."},{"key":"ref_23","doi-asserted-by":"crossref","unstructured":"Hrube\u0161, P., Langr, M., and Purkr\u00e1bkov\u00e1, Z. (2024, January 23\u201324). Review of Data Governance Approaches in the Field of Transportation Domain. Proceedings of the 2024 Smart City Symposium Prague (SCSP), Prague, Czech Republic.","DOI":"10.1109\/SCSP61506.2024.10552682"},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Si, S., Xiong, W., and Che, X. (2023). Data Quality Analysis and Improvement: A Case Study of a Bus Transportation System. Appl. Sci., 13.","DOI":"10.3390\/app131911020"},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Remoundou, K., Alexakis, T., Peppes, N., Demestichas, K., and Adamopoulou, E. (2022). A Quality Control Methodology for Heterogeneous Vehicular Data Streams. Sensors, 22.","DOI":"10.3390\/s22041550"},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Galarus, D., Turnbull, I., and Campbell, S. (2019, January 14\u201315). Timely, Reliable: A High Standard and Elusive Goal for Traveler Information Data Quality. Proceedings of the 2019 Future of Information and Communication Conference, San Francisco, CA, USA.","DOI":"10.1007\/978-3-030-12388-8_41"},{"key":"ref_27","unstructured":"Liu, Z., Li, L., Wang, Y., Lin, H., Liu, Z., He, L., and Wang, J. (2024). Controllable Traffic Simulation through Llm-Guided Hierarchical Chain-of-Thought Reasoning. arXiv."},{"key":"ref_28","doi-asserted-by":"crossref","unstructured":"Masri, S., Ashqar, H.I., and Elhenawy, M. (2025). Large Language Models (LLMs) as Traffic Control Systems at Urban Intersections: A New Paradigm. Vehicles, 7.","DOI":"10.3390\/vehicles7010011"},{"key":"ref_29","doi-asserted-by":"crossref","unstructured":"Masri, S., Ashqar, H.I., and Elhenawy, M. (2024). Leveraging Large Language Models (LLMs) for Traffic Management at Urban Intersections: The Case of Mixed Traffic Scenarios. arXiv.","DOI":"10.3390\/vehicles7010011"},{"key":"ref_30","doi-asserted-by":"crossref","unstructured":"Cheng, Q., Jiao, X., Yang, M., Yang, M., Jiang, K., and Yang, D. (2024, January 2\u20136). Advancing Autonomous Driving Safety Through LLM Enhanced Trajectory Prediction. Proceedings of the Advanced Vehicle Control Symposium, Milan, Italy.","DOI":"10.1007\/978-3-031-70392-8_71"},{"key":"ref_31","doi-asserted-by":"crossref","unstructured":"de Zarz\u00e0, I., de Curt\u00f2, J., Roig, G., and Calafate, C.T. (2023). LLM Multimodal Traffic Accident Forecasting. Sensors, 23.","DOI":"10.3390\/s23229225"},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Alhadidi, T., Jaber, A., Jaradat, S., Ashqar, H.I., and Elhenawy, M. (2024). Object Detection Using Oriented Window Learning Vi-Sion Transformer: Roadway Assets Recognition. arXiv.","DOI":"10.1007\/978-3-031-82377-0_41"},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"104413","DOI":"10.1016\/j.rineng.2025.104413","article-title":"Investigating Patterns of Freeway Crashes in Jordan: Findings from Text Mining Approach","volume":"26","author":"Jaradat","year":"2025","journal-title":"Results Eng."},{"key":"ref_34","unstructured":"Huang, X., Feng, Y., and Zhang, Z. (2024, January 27\u201329). Crash Report Generation Using ChatGPT: A Novel Approach for Automated Accident Reporting. Proceedings of the 2024 IEEE 4th International Conference on Electronic Information Engineering and Computer Science (EIECS), Yanji, China."},{"key":"ref_35","doi-asserted-by":"crossref","unstructured":"Pendyala, R., and Hall, S. (2024). Explaining Misinformation Detection Using Large Language Models. Electronics, 13.","DOI":"10.20944\/preprints202404.1513.v1"},{"key":"ref_36","doi-asserted-by":"crossref","first-page":"e00028824","DOI":"10.1590\/0102-311xen028824","article-title":"Comparative Analysis of BERT-Based and Generative Large Language Models for Detecting Suicidal Ideation: A Performance Evaluation Study","volume":"40","author":"Oliveira","year":"2024","journal-title":"Cad. Saude Publica"},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"157","DOI":"10.1162\/coli_a_00464","article-title":"Annotation Error Detection: Analyzing the Past and Present for a More Coherent Future","volume":"49","author":"Klie","year":"2022","journal-title":"Comput. Linguist."},{"key":"ref_38","first-page":"e60044","article-title":"Utilizing Large Language Models for Enhanced Clinical Trial Matching: A Study on Automation in Patient Screening","volume":"16","author":"Beattie","year":"2024","journal-title":"Cureus"},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"107523","DOI":"10.1016\/j.infsof.2024.107523","article-title":"Fine-Tuning and Prompt Engineering for Large Language Models-Based Code Review Automation","volume":"175","author":"Pornprasit","year":"2024","journal-title":"Inf. Softw. Technol."},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"100210","DOI":"10.1016\/j.caeai.2024.100210","article-title":"Fine-Tuning ChatGPT for Automatic Scoring","volume":"6","author":"Latif","year":"2024","journal-title":"Comput. Educ. Artif. Intell."},{"key":"ref_41","doi-asserted-by":"crossref","unstructured":"Davis, J., and Goadrich, M. (2006, January 25\u201329). The Relationship between Precision-Recall and ROC Curves. Proceedings of the 23rd International Conference on Machine Learning, Pittsburgh, PA, USA.","DOI":"10.1145\/1143844.1143874"},{"key":"ref_42","doi-asserted-by":"crossref","unstructured":"Saito, T., and Rehmsmeier, M. (2015). The Precision-Recall Plot Is More Informative than the ROC Plot When Evaluating Binary Classifiers on Imbalanced Datasets. PLoS ONE, 10.","DOI":"10.1371\/journal.pone.0118432"},{"key":"ref_43","unstructured":"Missouri State Highway Patrol (MSHP) (2022). Online Crash Report Search."},{"key":"ref_44","unstructured":"Mumtarin, M., Chowdhury, M.S., and Wood, J. (2023). Large Language Models in Analyzing Crash Narratives: A Comparative Study of ChatGPT, Bard and GPT-4. arXiv."},{"key":"ref_45","unstructured":"Bhagat, S., Shihab, I.F., and Sharma, A. (2024). Accuracy Is Not Agreement: Expert-Aligned Evaluation of Crash Narrative Classification Models. arXiv."},{"key":"ref_46","unstructured":"Bucher, M.J.J., and Martini, M. (2024). Fine-Tuned \u201csmall\u201d LLMs (Still) Significantly Outperform Zero-Shot Generative AI Models in Text Classification. arXiv."},{"key":"ref_47","first-page":"116","article-title":"Scalable Parallel Simulator for Vehicular Collision Detection","volume":"8","author":"Grinberg","year":"2013","journal-title":"Int. J. Veh. Saf. Mobil. Technol."}],"container-title":["Algorithms"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1999-4893\/18\/6\/317\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,9]],"date-time":"2025-10-09T17:41:11Z","timestamp":1760031671000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1999-4893\/18\/6\/317"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,5,27]]},"references-count":47,"journal-issue":{"issue":"6","published-online":{"date-parts":[[2025,6]]}},"alternative-id":["a18060317"],"URL":"https:\/\/doi.org\/10.3390\/a18060317","relation":{},"ISSN":["1999-4893"],"issn-type":[{"value":"1999-4893","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,5,27]]}}}