{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,18]],"date-time":"2026-08-18T05:14:54Z","timestamp":1787030094662,"version":"build-2736575974"},"reference-count":45,"publisher":"MDPI AG","issue":"2","license":[{"start":{"date-parts":[[2026,2,18]],"date-time":"2026-02-18T00:00:00Z","timestamp":1771372800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100016047","name":"Ministry of Education, Science and Technological Development of the Republic of Serbia, as well as by the Science Fund of the Republic of Serbia","doi-asserted-by":"publisher","award":["7373"],"award-info":[{"award-number":["7373"]}],"id":[{"id":"10.13039\/501100016047","id-type":"DOI","asserted-by":"publisher"}]},{"name":"European Union\u2014NextGenerationEU"},{"name":"Institute for Medical Research and Occupational Health, Zagreb, Croatia","award":["643-02\/23-01\/00016"],"award-info":[{"award-number":["643-02\/23-01\/00016"]}]},{"name":"Institute for Medical Research and Occupational Health, Zagreb, Croatia","award":["533-03-23-0006"],"award-info":[{"award-number":["533-03-23-0006"]}]},{"name":"City of Zagreb"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Toxics"],"abstract":"<jats:p>We investigated benzene variability in an urban environment using an interpretable, setting-based artificial intelligence framework. A seven-year dataset (2017\u20132023) of hourly pollutant concentrations (benzene, NO2, SO2, CO, O3) measured in Zagreb (Croatia) was analyzed, as were meteorological variables. Multiple-ensemble decision tree models were developed, with hyperparameters optimized using metaheuristic algorithms. The best-performing model, Extra Trees optimized by the Sine Cosine Algorithm, achieved an R2 of 0.87. Model interpretation employed Shapley additive explanations (SHAP), followed by PaCMAP embedding and HDBSCAN clustering to identify coherent environmental settings. Seven settings (C0\u2013C6) and one residual group were identified, representing pollution-enhancing, suppressing, and transitional regimes. Two settings dominated benzene extremes. C6 reflected winter stagnation, characterized by strong combustion influence (CO contribution of 11.9%), shallow boundary layers (~290 m), weak winds, and high humidity. C4 represented a synoptic stability regime with enhanced heat fluxes and diminished after the COVID-19 period, consistent with altered anthropogenic activity. Low-benzene settings (C0, C1, C3) were associated with stronger mixing and higher oxidizing capacity, while transitional settings (C2, C5) reflected moderate conditions. Overall, the results show that a small number of environmental settings governed the benzene extremes, providing a transferable and interpretable framework for air quality assessment and policy support.<\/jats:p>","DOI":"10.3390\/toxics14020181","type":"journal-article","created":{"date-parts":[[2026,2,18]],"date-time":"2026-02-18T09:22:46Z","timestamp":1771406566000},"page":"181","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":3,"title":["Uncovering Benzene Pollution Patterns Using an Interpretable, Setting-Aware Artificial Intelligence Approach"],"prefix":"10.3390","volume":"14","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-6041-2299","authenticated-orcid":false,"given":"Ivan","family":"Be\u0161li\u0107","sequence":"first","affiliation":[{"name":"Institute for Medical Research and Occupational Health, Ksaverska cesta 2, P.O. Box 291, 10001 Zagreb, Croatia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6938-6974","authenticated-orcid":false,"given":"Timea","family":"Bezdan","sequence":"additional","affiliation":[{"name":"Faculty of Informatics and Computing, Singidunum University, Danijelova 32, 11000 Belgrade, Serbia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8657-423X","authenticated-orcid":false,"given":"Gordana","family":"Jovanovi\u0107","sequence":"additional","affiliation":[{"name":"Institute of Physics Belgrade, National Institute of the Republic of Serbia, Pregrevica 118, 11080 Belgrade, Serbia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3474-9883","authenticated-orcid":false,"given":"Silvije","family":"Davila","sequence":"additional","affiliation":[{"name":"Institute for Medical Research and Occupational Health, Ksaverska cesta 2, P.O. Box 291, 10001 Zagreb, Croatia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5155-1847","authenticated-orcid":false,"given":"Gordana","family":"Pehnec","sequence":"additional","affiliation":[{"name":"Institute for Medical Research and Occupational Health, Ksaverska cesta 2, P.O. Box 291, 10001 Zagreb, Croatia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2382-4734","authenticated-orcid":false,"given":"Snje\u017eana","family":"Herceg Romani\u0107","sequence":"additional","affiliation":[{"name":"Institute for Medical Research and Occupational Health, Ksaverska cesta 2, P.O. Box 291, 10001 Zagreb, Croatia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5293-9533","authenticated-orcid":false,"given":"Andreja","family":"Stoji\u0107","sequence":"additional","affiliation":[{"name":"Institute of Physics Belgrade, National Institute of the Republic of Serbia, Pregrevica 118, 11080 Belgrade, Serbia"},{"name":"Environment and Sustainable Development Studies, Singidunum University, Danijelova 32, 11000 Belgrade, Serbia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8287-4136","authenticated-orcid":false,"given":"Mirjana","family":"Peri\u0161i\u0107","sequence":"additional","affiliation":[{"name":"Institute of Physics Belgrade, National Institute of the Republic of Serbia, Pregrevica 118, 11080 Belgrade, Serbia"},{"name":"Environment and Sustainable Development Studies, Singidunum University, Danijelova 32, 11000 Belgrade, Serbia"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,2,18]]},"reference":[{"key":"ref_1","unstructured":"World Health Organization (2025). WHO Strategic Approach on Air Quality, Energy Access and Health, World Health Organization. Available online: https:\/\/iris.who.int\/server\/api\/core\/bitstreams\/42ce49d3-1501-474c-a860-bf71752a2113\/content."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"1907","DOI":"10.1016\/S0140-6736(17)30505-6","article-title":"Estimates and 25-year trends of the global burden of disease attributable to ambient air pollution: An analysis of data from the Global Burden of Diseases Study 2015","volume":"389","author":"Cohen","year":"2017","journal-title":"Lancet"},{"key":"ref_3","doi-asserted-by":"crossref","unstructured":"Nie, Y., Yan, Y., Ji, Y., Gao, R., Ren, Y., Bi, F., Shang, F., Li, J., Chu, W., and Li, H. (2025). Assessing the PM2.5\u2013O3 Correlation and Unraveling Their Drivers in Urban Environment: Insights from the Bohai Bay Region, China. Atmosphere, 16.","DOI":"10.3390\/atmos16050512"},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Radi\u0107, N., Peri\u0161i\u0107, M., Jovanovi\u0107, G., Bezdan, T., Stani\u0161i\u0107, S., Stani\u0107, N., and Stoji\u0107, A. (2025). An AI-based framework for characterizing the atmospheric fate of air pollutants within diverse environmental settings. Atmosphere, 16.","DOI":"10.3390\/atmos16020231"},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"6975","DOI":"10.5194\/acp-25-6975-2025","article-title":"Significant contributions of biomass burning to PM2.5-bound aromatic compounds","volume":"25","author":"Ren","year":"2025","journal-title":"Atmos. Chem. Phys."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"2459","DOI":"10.5194\/acp-25-2459-2025","article-title":"Vertical changes in volatile organic compounds (VOCs) and impacts on photochemical ozone formation","volume":"25","author":"Li","year":"2025","journal-title":"Atmos. Chem. Phys."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"7699","DOI":"10.5194\/acp-24-7699-2024","article-title":"Evaluation of modelled versus observed non-methane volatile organic compounds at European Monitoring and Evaluation Programme sites in Europe","volume":"24","author":"Ge","year":"2024","journal-title":"Atmos. Chem. Phys."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"106312","DOI":"10.1016\/j.envsoft.2024.106312","article-title":"Application of artificial intelligence in air pollution monitoring and forecasting: A systematic review","volume":"185","author":"Chadalavada","year":"2025","journal-title":"Environ. Model. Softw."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"464","DOI":"10.1007\/s11270-025-08122-8","article-title":"Air Quality Forecasting Using Machine Learning: Comparative Analysis and Ensemble Strategies for Enhanced Prediction","volume":"236","author":"Alpsalaz","year":"2025","journal-title":"Water Air Soil Pollut."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"195","DOI":"10.1038\/s41586-019-0912-1","article-title":"Deep learning and process understanding for data-driven Earth system science","volume":"566","author":"Reichstein","year":"2019","journal-title":"Nature"},{"key":"ref_11","unstructured":"Lundberg, S.M., and Lee, S.I. (2017). A unified approach to interpreting model predictions. Adv. Neural Inf. Process. Syst., 30."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"205","DOI":"10.21105\/joss.00205","article-title":"HDBSCAN: Hierarchical density-based clustering","volume":"2","author":"McInnes","year":"2017","journal-title":"J. Open Source Softw."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"113076","DOI":"10.1016\/j.asoc.2025.113076","article-title":"Exploiting synthetic data generation to enhance pollution prediction","volume":"175","author":"Santa","year":"2025","journal-title":"Appl. Soft Comput."},{"key":"ref_14","unstructured":"(2012). Ambient Air\u2014Standard Method for the Measurement of the Concentration of Nitrogen Dioxide and Nitrogen Monoxide by Chemiluminescence (Standard No. EN 14211:2012)."},{"key":"ref_15","unstructured":"(2012). Ambient Air\u2014Standard Method for the Measurement of the Concentration of Sulphur Dioxide by Ultraviolet Fluorescence (Standard No. EN 14212:2012)."},{"key":"ref_16","unstructured":"(2012). Ambient Air\u2014Standard Method for the Measurement of the Concentration of Ozone by Ultraviolet Photometry (Standard No. EN 14625:2012)."},{"key":"ref_17","unstructured":"(2012). Ambient Air\u2014Standard Method for the Measurement of the Concentration of Carbon Monoxide by Non-Dispersive Infrared Spectroscopy (Standard No. EN 14626:2012)."},{"key":"ref_18","unstructured":"(2015). Ambient Air\u2014Standard Method for the Measurement of Benzene Concentrations\u2014Part 3: Automated Pumped Sampling with in Situ Gas Chromatography (Standard No. EN 14662-3:2015)."},{"key":"ref_19","unstructured":"(2017). General Requirements for the Competence of Testing and Calibration Laboratories. 3rd Edition (Standard No. ISO\/IEC 17025:2017(E))."},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Dietterich, T.G. (2000). Ensemble methods in machine learning. International Workshop on Multiple Classifier Systems, Springer.","DOI":"10.1007\/3-540-45014-9_1"},{"key":"ref_21","unstructured":"Python Software Foundation (2025, December 15). Python Language Reference, Available online: https:\/\/www.python.org\/."},{"key":"ref_22","first-page":"1","article-title":"Understanding how dimension reduction tools work: An empirical approach to deciphering t-SNE, UMAP, TriMap, and PaCMAP for data visualization","volume":"22","author":"Wang","year":"2021","journal-title":"J. Mach. Learn. Res."},{"key":"ref_23","unstructured":"R Core Team (2023). R: A Language and Environment for Statistical Computing, R Foundation for Statistical Computing. Available online: https:\/\/www.R-project.org\/."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"3","DOI":"10.1007\/s10994-006-6226-1","article-title":"Extremely randomized trees","volume":"63","author":"Geurts","year":"2006","journal-title":"Mach. Learn."},{"key":"ref_25","first-page":"2825","article-title":"Scikit-learn: Machine learning in Python","volume":"12","author":"Pedregosa","year":"2011","journal-title":"J. Mach. Learn. Res."},{"key":"ref_26","unstructured":"Ke, G., Meng, Q., Finley, T., Wang, T., Chen, W., Ma, W., Ye, Q., and Liu, T.Y. (2017). LightGBM: A highly efficient gradient boosting decision tree. Adv. Neural Inf. Process. Syst., 30, Available online: https:\/\/proceedings.neurips.cc\/paper_files\/paper\/2017\/file\/6449f44a102fde848669bdd9eb6b76fa-Paper.pdf."},{"key":"ref_27","doi-asserted-by":"crossref","unstructured":"Chen, T., and Guestrin, C. (2016, January August). XGBoost: A Scalable Tree Boosting System. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD\u201916), San Francisco, CA, USA.","DOI":"10.1145\/2939672.2939785"},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"849","DOI":"10.1016\/j.future.2019.02.028","article-title":"Harris hawks optimization: Algorithm and applications","volume":"97","author":"Heidari","year":"2019","journal-title":"FGCS"},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"120","DOI":"10.1016\/j.knosys.2015.12.022","article-title":"SCA: A sine cosine algorithm for solving optimization problems","volume":"96","author":"Mirjalili","year":"2016","journal-title":"Knowl.-Based Syst."},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"133154","DOI":"10.1016\/j.chemosphere.2021.133154","article-title":"The PM2.5-bound polycyclic aromatic hydrocarbon behavior in indoor and outdoor environments, part II: Explainable prediction of benzo [a] pyrene levels","volume":"289","year":"2022","journal-title":"Chemosphere"},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"467","DOI":"10.1007\/s11869-020-00950-3","article-title":"Influence of lockdown caused by the COVID-19 pandemic on air pollution and carcinogenic content of particulate matter observed in Croatia","volume":"14","author":"Godec","year":"2021","journal-title":"Air Qual. Atmos. Health"},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Lovri\u0107, M., Antunovi\u0107, M., \u0160uni\u0107, I., Vukovi\u0107, M., Kecorius, S., Kr\u00f6ll, M., Be\u0161li\u0107, I., Godec, R., Pehnec, G., and Geiger, B. (2022). Machine learning and meteorological normalization for assessment of particulate matter changes during the COVID-19 lockdown in Zagreb, Croatia. Int. J. Environ. Res. Public Health, 19.","DOI":"10.3390\/ijerph19116937"},{"key":"ref_33","unstructured":"Copernicus Atmosphere Monitoring Service (CAMS) (2025, December 15). Repeated Summer Heatwaves Trigger New Ozone-Level Peaks Throughout Europe. Available online: https:\/\/atmosphere.copernicus.eu\/repeated-summer-heatwaves-trigger-new-ozone-level-peaks-throughout-europe."},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"36640","DOI":"10.1007\/s11356-021-13295-w","article-title":"Assessment of ozone concentration data from the northern Zagreb area, Croatia, for the period from 2003 to 2016","volume":"28","author":"Pehnec","year":"2021","journal-title":"Environ. Sci. Pollut. Res."},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"2911","DOI":"10.5194\/acp-20-2911-2020","article-title":"How much does traffic contribute to benzene and polycyclic aromatic hydrocarbon air pollution? Results from a high-resolution North American air quality model centred on Toronto, Canada","volume":"20","author":"Whaley","year":"2020","journal-title":"Atmos. Chem. Phys."},{"key":"ref_36","doi-asserted-by":"crossref","first-page":"104675","DOI":"10.1016\/j.atmosres.2019.104675","article-title":"On the association between characteristics of the atmospheric boundary layer and air pollution concentrations","volume":"231","author":"Dayan","year":"2020","journal-title":"Atmos. Res."},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"493","DOI":"10.1016\/j.scitotenv.2019.03.097","article-title":"Impacts of the near-surface urban boundary layer structure on PM2.5 concentrations in Beijing during winter","volume":"669","author":"Wang","year":"2019","journal-title":"Sci. Total Environ."},{"key":"ref_38","doi-asserted-by":"crossref","unstructured":"Calvert, J.G., Atkinson, R., Becker, K.H., Kamens, R.M., Seinfeld, J.H., Wallington, T.J., and Yarwood, G. (2002). Aerosol generation in atmospheric oxidation of aromatic hydrocarbons. The mechanisms of Atmospheric Oxidation of Aromatic Hydrocarbons, Oxford University Press. Chapter VIII.","DOI":"10.1093\/oso\/9780195146288.001.0001"},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"4218","DOI":"10.1021\/cr500648z","article-title":"Heterogeneous photochemistry in the atmosphere","volume":"115","author":"George","year":"2015","journal-title":"Chem. Rev."},{"key":"ref_40","unstructured":"Seinfeld, J.H., and Pandis, S.N. (2016). Atmospheric Chemistry and Physics: From Air Pollution to Climate Change, Wiley. [3rd ed.]."},{"key":"ref_41","doi-asserted-by":"crossref","first-page":"3856","DOI":"10.1021\/acs.chemrev.5b00089","article-title":"Chemistry and the Linkages between Air Quality and Climate Change","volume":"115","author":"Monks","year":"2015","journal-title":"Chem. Rev."},{"key":"ref_42","doi-asserted-by":"crossref","first-page":"262","DOI":"10.1038\/s41612-024-00818-8","article-title":"Elucidating key factors in regulating budgets of ozone and its precursors in atmospheric boundary layer","volume":"7","author":"Song","year":"2024","journal-title":"npj Clim. Atmos. Sci."},{"key":"ref_43","doi-asserted-by":"crossref","unstructured":"Sanda, M., Dunea, D., Iordache, S., Pohoata, A., Glod-Lendvai, A.-M., and Onutu, I. (2023). A three-year analysis of toxic benzene levels and associated impact in Ploie\u015fti city, Romania. Toxics, 11.","DOI":"10.3390\/toxics11090748"},{"key":"ref_44","doi-asserted-by":"crossref","first-page":"100127","DOI":"10.1016\/j.envadv.2021.100127","article-title":"Source apportionment of volatile organic compounds, CO, SO2 and trace metals in a complex urban atmosphere","volume":"6","author":"Ahmed","year":"2021","journal-title":"Environ. Adv."},{"key":"ref_45","doi-asserted-by":"crossref","first-page":"9210","DOI":"10.1109\/JBHI.2024.3508466","article-title":"Reducing Pollution Health Impact with Air Quality Prediction Assisted by Mobility Data","volume":"29","author":"Santa","year":"2025","journal-title":"IEEE J. Biomed. Health Inform."}],"container-title":["Toxics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2305-6304\/14\/2\/181\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,2,18]],"date-time":"2026-02-18T09:33:25Z","timestamp":1771407205000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2305-6304\/14\/2\/181"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,2,18]]},"references-count":45,"journal-issue":{"issue":"2","published-online":{"date-parts":[[2026,2]]}},"alternative-id":["toxics14020181"],"URL":"https:\/\/doi.org\/10.3390\/toxics14020181","relation":{},"ISSN":["2305-6304"],"issn-type":[{"value":"2305-6304","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,2,18]]}}}