{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,3]],"date-time":"2026-02-03T00:56:51Z","timestamp":1770080211410,"version":"3.49.0"},"reference-count":61,"publisher":"MDPI AG","issue":"4","license":[{"start":{"date-parts":[[2025,11,10]],"date-time":"2025-11-10T00:00:00Z","timestamp":1762732800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Nitrogen"],"abstract":"<jats:p>This study aims to evaluate and compare the performance of Autoencoders (AEs) and Sparse Autoencoders (SAEs) in forecasting the next-hour concentration levels of various air pollutants\u2014specifically NO2(t + 1), PM10(t + 1), and SO2(t + 1)\u2014in the Bay of Algeciras, a highly complex region located in southern Spain. Hourly data related to air quality, meteorological conditions, and maritime traffic were collected from 2017 to 2019 across multiple monitoring stations distributed throughout the bay, enabling the analysis of diverse forecasting scenarios. The output variable was segmented into four distinct, non-overlapping quartiles (Q1\u2013Q4) to capture different concentration ranges. AE models demonstrated greater accuracy in predicting moderate pollution levels (Q2 and Q3), whereas SAE models achieved comparable performance at the lower and upper extremes (Q1 and Q4). The results suggest that stacking AE layers with varying degrees of sparsity\u2014culminating in a supervised output layer\u2014can enhance the model\u2019s ability to forecast pollutant concentration indices across all quartiles. Notably, Q4 predictions, representing peak concentrations, benefited from more complex SAE architectures, likely due to the increased difficulty associated with modelling extreme values.<\/jats:p>","DOI":"10.3390\/nitrogen6040101","type":"journal-article","created":{"date-parts":[[2025,11,10]],"date-time":"2025-11-10T08:57:55Z","timestamp":1762765075000},"page":"101","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":2,"title":["Air Pollution Forecasting Using Autoencoders: A Classification-Based Prediction of NO2, PM10, and SO2 Concentrations"],"prefix":"10.3390","volume":"6","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-1364-1678","authenticated-orcid":false,"given":"Mar\u00eda Inmaculada","family":"Rodr\u00edguez-Garc\u00eda","sequence":"first","affiliation":[{"name":"Department of Computer Science Engineering, Algeciras Higher School of Engineering (ETSIA), University of C\u00e1diz, 11202 Algeciras, Spain"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8478-6431","authenticated-orcid":false,"given":"Mar\u00eda Gema","family":"Carrasco-Garc\u00eda","sequence":"additional","affiliation":[{"name":"Department of Industrial and Civil Engineering, Algeciras Higher School of Engineering (ETSIA), University of C\u00e1diz, 11202 Algeciras, Spain"}]},{"given":"Paloma Roc\u00edo","family":"Cubillas Fern\u00e1ndez","sequence":"additional","affiliation":[{"name":"Department of Thermal Machines and Engines, Higher Technical School of Engineering of Algeciras (ETSIA), University of C\u00e1diz, 11202 Algeciras, Spain"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0185-3200","authenticated-orcid":false,"given":"Maria da Concei\u00e7ao","family":"Rodrigues Ribeiro","sequence":"additional","affiliation":[{"name":"Engineering Institute, Campus da Penha, University of Algarve, 8005-139 Faro, Portugal"},{"name":"CEAUL\u2014Centre of Statistics and Its Applications, Faculdade de Ci\u00eancias, Universidade de Lisboa, 1749-016 Lisboa, Portugal"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4803-7964","authenticated-orcid":false,"given":"Pedro J. S.","family":"Cardoso","sequence":"additional","affiliation":[{"name":"NOVA Laboratory for Computer Science and Informatics (NOVA LINCS), Instituto Superior de Engenharia, Campus da Penha, University of Algarve, 8005-139 Faro, Portugal"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4627-0252","authenticated-orcid":false,"given":"Ignacio. J.","family":"Turias","sequence":"additional","affiliation":[{"name":"Department of Computer Science Engineering, Algeciras Higher School of Engineering (ETSIA), University of C\u00e1diz, 11202 Algeciras, Spain"}]}],"member":"1968","published-online":{"date-parts":[[2025,11,10]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"161","DOI":"10.1186\/s40537-021-00548-1","article-title":"Air-pollution prediction in smart city, deep learning approach","volume":"8","author":"Bekkar","year":"2021","journal-title":"J. Big Data"},{"key":"ref_2","doi-asserted-by":"crossref","unstructured":"Agbehadji, I.E., and Obagbuwa, I.C. (2024). Systematic Review of Machine Learning and Deep Learning Techniques for Spatiotemporal Air Quality Prediction. Atmosphere, 15.","DOI":"10.3390\/atmos15111352"},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"8439","DOI":"10.5194\/gmd-15-8439-2022","article-title":"Development of an LSTM broadcasting deep-learning framework for regional air pollution forecast improvement","volume":"15","author":"Sun","year":"2022","journal-title":"Geosci. Model. Dev."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"8512","DOI":"10.1021\/es071686z","article-title":"Mortality from ship emissions: A global assessment","volume":"41","author":"Corbett","year":"2007","journal-title":"Environ. Sci. Technol."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"4735","DOI":"10.1016\/j.atmosenv.2009.04.059","article-title":"Transport impacts on atmosphere and climate: Shipping","volume":"44","author":"Eyring","year":"2010","journal-title":"Atmos. Environ."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"1449","DOI":"10.5194\/acp-12-1449-2012","article-title":"Modelling the climate impact of road transport, maritime shipping and aviation over the period 1860\u20132100 with an AOGCM","volume":"12","author":"Cariolle","year":"2012","journal-title":"Atmos. Chem. Phys."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"6801","DOI":"10.5194\/acp-14-6801-2014","article-title":"The climate impact of ship NOx emissions: An improved estimate accounting for plume chemistry","volume":"14","author":"Holmes","year":"2014","journal-title":"Atmos. Chem. Phys."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"61","DOI":"10.1016\/j.atmosres.2012.12.009","article-title":"Contribution of ship emissions to the fine particulate in the community near an international port in Hong Kong","volume":"124","author":"Yau","year":"2013","journal-title":"Atmos. Res."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"105447","DOI":"10.1016\/j.marpol.2022.105447","article-title":"Air pollutant emission calculation and shipping costs: The case of yacht transport between the North Sea and Mediterranean ports","volume":"148","author":"Guidi","year":"2023","journal-title":"Mar. Policy"},{"key":"ref_10","first-page":"4183","article-title":"Models-3 Community Multiscale Air Quality (CMAQ) model aerosol component. Model description","volume":"108","author":"Binkowski","year":"2003","journal-title":"Environ. Sci."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"31","DOI":"10.1186\/1745-6673-6-31","article-title":"Ships, ports and particulate air pollution\u2014An analysis of recent studies","volume":"6","author":"Mueller","year":"2011","journal-title":"J. Occup. Med. Toxicol."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"107284","DOI":"10.1016\/j.envint.2022.107284","article-title":"Short-term NO2 exposure and cognitive and mental health: A panel study based on a citizen science project in Barcelona, Spain","volume":"164","author":"Gignac","year":"2022","journal-title":"Environ. Int."},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Lili, L., Zhu, J., Ye, G., and Feng, X. (2018). Development of Green Ports with the Con-sideration of Coastal Wave Energy. Sustainability, 10.","DOI":"10.3390\/su10114270"},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"243","DOI":"10.1016\/j.trd.2019.05.003","article-title":"Reviewing tools and technologies for sustainable ports: Does research enable decision making in ports?","volume":"72","author":"Ystmark","year":"2019","journal-title":"Transp. Res. Part. D Transp. Environ."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"117674","DOI":"10.1016\/j.atmosenv.2020.117674","article-title":"Contributions of traffic and shipping emissions to city-scale NOx and PM2.5 exposures in Hamburg","volume":"237","author":"Ramacher","year":"2020","journal-title":"Atmos. Environ."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"46","DOI":"10.1016\/j.apr.2020.08.012","article-title":"Chemical characterization of PM2.5 and PM2.5\u201310 samples collected in urban site in Mediterranean coast of Turkey","volume":"12","author":"Tepe","year":"2021","journal-title":"Atmos. Pollut. Res."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"105536","DOI":"10.1016\/j.marpol.2023.105536","article-title":"A novel FMEA approach for risk assessment of air pollution from ships","volume":"150","author":"Ceylan","year":"2023","journal-title":"Mar. Policy"},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Rapalis, P., \u0160ilas, G., and \u017daglinskis, J. (2022). Ship Air Pollution Estimation by AIS Da-ta: Case Port of Klaipeda. J. Mar. Sci. Eng., 10.","DOI":"10.3390\/jmse10121950"},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"1","DOI":"10.11591\/ijai.v12.i1.pp1-11","article-title":"Automatic identification system-based trajectory clustering framework to identify vessel movement pattern","volume":"12","author":"Widyantara","year":"2023","journal-title":"IAES Int. J. Artif. Intell."},{"key":"ref_20","first-page":"71","article-title":"Statistical Analysis for Model-ling Ship Operation Processes in Ports","volume":"62","author":"Nas","year":"2020","journal-title":"Sci. J. Marit. Univ. Szczec."},{"key":"ref_21","first-page":"169","article-title":"A comprehensive inventory of ship traffic exhaust emissions","volume":"69","author":"Jalkanen","year":"2012","journal-title":"J. Atmos. Chem."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"1999","DOI":"10.1007\/s00477-021-01992-4","article-title":"A comparison of ranking filter methods applied to the estimation of NO2 concentrations in the Bay of Algeciras (Spain)","volume":"35","author":"Urda","year":"2021","journal-title":"Stoch. Environ. Res. Risk Assess."},{"key":"ref_23","doi-asserted-by":"crossref","unstructured":"Rodr\u00edguez-Garc\u00eda, M.I., Carrasco-Garc\u00eda, M.G., Ribeiro, M.d.C.R., Gonz\u00e1lez-Enrique, J., Ruiz-Aguilar, J.J., and Turias, I.J. (2024). Air Pollution PM10 Forecasting Maps in the Maritime Area of the Bay of Algeciras (Spain). J. Mar. Sci. Eng., 12.","DOI":"10.3390\/jmse12030397"},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"4359","DOI":"10.1007\/s00477-023-02512-2","article-title":"Forecasting air pollutants using classification models: A case study in the Bay of Algeciras (Spain)","volume":"37","author":"Turias","year":"2023","journal-title":"Stoch. Environ. Res. Risk Assess."},{"key":"ref_25","first-page":"58","article-title":"Air Pollution Modelling with Deep Learning: A Review","volume":"1","author":"Akin","year":"2018","journal-title":"Int. J. Environ. Pollut. Environ. Model."},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"22408","DOI":"10.1007\/s11356-016-7812-9","article-title":"Deep learning architecture for air quality predictions","volume":"23","author":"Li","year":"2016","journal-title":"Environ. Sci. Pollut. Res."},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"104600","DOI":"10.1016\/j.envsoft.2019.104600","article-title":"Constructing a PM2.5 concentration prediction model by combining autoencoder with Bi-LSTM neural networks","volume":"124","author":"Zhang","year":"2020","journal-title":"Environ. Model. Softw."},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"399","DOI":"10.1007\/s40726-020-00159-z","article-title":"Deep Learning for Air Quality Forecasts: A Review","volume":"6","author":"Liao","year":"2020","journal-title":"Curr. Pollut. Rep."},{"key":"ref_29","doi-asserted-by":"crossref","unstructured":"Xayasouk, T., Lee, H., and Lee, G. (2020). Air Pollution Prediction Using Long Short-Term Memory (LSTM) and Deep Autoencoder (DAE) Models. Sustainability, 12.","DOI":"10.3390\/su12062570"},{"key":"ref_30","unstructured":"Arcucci, R., Mottet, L., Guo, Y., and Pain, C. (2021). Adversarial autoencoders and adversarial LSTM for improved forecasts of urban air pollution simulations. arXiv."},{"key":"ref_31","doi-asserted-by":"crossref","unstructured":"Mengara, A.G., Park, E., Jang, J., and Yoo, Y. (2022). Attention-Based Distributed Deep Learning Model for Air Quality Forecasting. Sustainability, 14.","DOI":"10.3390\/su14063269"},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"109","DOI":"10.34306\/att.v5i1Sp.337","article-title":"Image-based Air Quality Prediction using Convolutional Neural Networks and Machine Learning","volume":"5","author":"Hardini","year":"2023","journal-title":"Aptisi Trans. Technopreneurship"},{"key":"ref_33","doi-asserted-by":"crossref","unstructured":"Rodr\u00edguez-Garc\u00eda, M.I., Carrasco-Garc\u00eda, M.G., Gonz\u00e1lez-Enrique, J., Ruiz-Aguilar, J.J., and Turias, I.J. (2023). Long Short-Term Memory Approach for Short-Term Air Quality Forecasting in the Bay of Algeciras (Spain). Sustainability, 15.","DOI":"10.3390\/su15065089"},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"436","DOI":"10.1038\/nature14539","article-title":"Deep learning","volume":"521","author":"LeCun","year":"2015","journal-title":"Nature"},{"key":"ref_35","doi-asserted-by":"crossref","unstructured":"Baldorj, B., Tsagaan, M., Sereeter, L., and Bulkhbai, A. (2022). Embedded Generative Air Pollution Model with Variational Autoencoder and Environmental Factor Effect in Ulaanbaatar City. Atmosphere, 13.","DOI":"10.3390\/atmos13010071"},{"key":"ref_36","doi-asserted-by":"crossref","first-page":"100872","DOI":"10.1016\/j.uclim.2021.100872","article-title":"Temporal convolutional denoising autoencoder network for air pollution prediction with missing values","volume":"38","year":"2021","journal-title":"Urban. Clim."},{"key":"ref_37","doi-asserted-by":"crossref","unstructured":"Wei, Y., Jang-Jaccard, J., Xu, W., Sabrina, F., Camtepe, S., and Boulic, M. (2022). LSTM-Autoencoder based Anomaly Detection for Indoor Air Quality Time Series Data. arXiv.","DOI":"10.1109\/JSEN.2022.3230361"},{"key":"ref_38","doi-asserted-by":"crossref","unstructured":"Mengara Mengara, A.G., Kim, Y., Yoo, Y., and Ahn, J. (2020). Distributed Deep Features Extraction Model for Air Quality Forecasting. Sustainability, 12.","DOI":"10.3390\/su12198014"},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"71","DOI":"10.2495\/AIR180071","article-title":"Air pollution prediction system using deep learning","volume":"230","author":"Xayasouk","year":"2018","journal-title":"WIT Trans. Ecol. Environ."},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"1","DOI":"10.31436\/iiumej.v26i1.2818","article-title":"Autoencoder Artificial Neural Network Model for Air Pollution Index Prediction","volume":"26","author":"Basir","year":"2025","journal-title":"IIUM Eng. J."},{"key":"ref_41","doi-asserted-by":"crossref","first-page":"113542","DOI":"10.1016\/j.envpol.2019.113542","article-title":"Impact of harbour activities on local air quality: A review","volume":"257","author":"Sorte","year":"2020","journal-title":"Environ. Pollut."},{"key":"ref_42","doi-asserted-by":"crossref","first-page":"863","DOI":"10.1016\/j.matpr.2020.07.141","article-title":"Air quality indices: A review of methods to interpret air quality status","volume":"34","author":"Suman","year":"2021","journal-title":"Mater. Today Proc."},{"key":"ref_43","doi-asserted-by":"crossref","first-page":"111681","DOI":"10.1016\/j.jenvman.2020.111681","article-title":"A review of current air quality indexes and improvements under the multi-contaminant air pollution exposure","volume":"279","author":"Tan","year":"2021","journal-title":"J. Environ. Manag."},{"key":"ref_44","doi-asserted-by":"crossref","first-page":"324","DOI":"10.1007\/s10661-022-09896-8","article-title":"A critical evaluation of air quality index models (1960\u20132021)","volume":"194","author":"Kumar","year":"2022","journal-title":"Environ. Monit. Assess."},{"key":"ref_45","doi-asserted-by":"crossref","first-page":"165","DOI":"10.1007\/s11157-010-9227-2","article-title":"Air quality indices: A review","volume":"10","author":"Plaia","year":"2011","journal-title":"Rev. Environ. Sci. Biotechnol."},{"key":"ref_46","doi-asserted-by":"crossref","unstructured":"Ting, K.M. (2010). Confusion matrix. Encyclopedia of Machine Learning and Data Mining, Springer.","DOI":"10.1007\/978-0-387-30164-8_157"},{"key":"ref_47","doi-asserted-by":"crossref","first-page":"533","DOI":"10.1038\/323533a0","article-title":"Learning representations by backpropagating errors","volume":"323","author":"Rumelhart","year":"1986","journal-title":"Nature"},{"key":"ref_48","doi-asserted-by":"crossref","first-page":"136","DOI":"10.1016\/j.neunet.2012.04.011","article-title":"Complex-valued autoencoders","volume":"33","author":"Baldi","year":"2012","journal-title":"Neural Netw."},{"key":"ref_49","unstructured":"Makhzani, A., and Frey, B. (2014, January 14\u201316). k-Sparse autoencoders. Proceedings of the 2nd International Conference on Learning Representations, ICLR Conference Track Proceedings, Banff, AB, Canada."},{"key":"ref_50","unstructured":"Tino, P., Benuskova, L., and Sperduti, A. (2025). Artificial neural network models. Springer Handbook of Computational Intelligence, Springer."},{"key":"ref_51","unstructured":"Kingma, D.P., and Welling, M. (2013). Auto-Encoding Variational Bayes. arXiv."},{"key":"ref_52","first-page":"281","article-title":"Random search for hyperparameter optimization","volume":"13","author":"Bergstra","year":"2012","journal-title":"J. Mach. Learn. Res."},{"key":"ref_53","doi-asserted-by":"crossref","first-page":"50","DOI":"10.1016\/j.gloei.2023.02.005","article-title":"Towards sparse matrix operations: Graph database approach for power grid computation","volume":"6","author":"Li","year":"2023","journal-title":"Glob. Energy Interconnect."},{"key":"ref_54","doi-asserted-by":"crossref","first-page":"170","DOI":"10.1136\/bmj.310.6973.170","article-title":"Multiple significance tests: The Bonferroni method","volume":"310","author":"Bland","year":"1995","journal-title":"BMJ"},{"key":"ref_55","unstructured":"Nasabpour, S., Salajegheh, A., Hassan Khosravi, H., Nasiti, A., and Ranjbar, A. (2023). Prediction of hourly PM10 concentration through a hybrid deep learning-based method. Res. Sq."},{"key":"ref_56","doi-asserted-by":"crossref","first-page":"6631614","DOI":"10.1155\/2021\/6631614","article-title":"A Novel Method for Regional NO2 Concentration Prediction Using Discrete Wavelet Transform and an LSTM Network","volume":"2021","author":"Liu","year":"2021","journal-title":"Comput. Intell. Neurosci."},{"key":"ref_57","doi-asserted-by":"crossref","first-page":"101051","DOI":"10.1016\/j.uclim.2021.101051","article-title":"Prediction of SO2 and PM10 air pollutants using a deep learning-based recurrent neural network: Case of industrial city Sakarya","volume":"41","author":"Kurnaz","year":"2022","journal-title":"Urban. Clim."},{"key":"ref_58","doi-asserted-by":"crossref","unstructured":"Yi, X., Zhang, J., Wang, Z., Li, T., and Zheng, Y. (2018, January 19\u201323). Deep Distributed Fusion Network for Air Quality Prediction. Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining (KDD \u201818). Association for Computing Machinery, New York, NY, USA.","DOI":"10.1145\/3219819.3219822"},{"key":"ref_59","doi-asserted-by":"crossref","first-page":"e2021MS002806","DOI":"10.1029\/2021MS002806","article-title":"Spatio-Temporal Hourly and Daily Ozone Forecasting in China Using a Hybrid Machine Learning Model: Autoencoder and Generative Adversarial Networks","volume":"14","author":"Cheng","year":"2022","journal-title":"J. Adv. Model. Earth Syst."},{"key":"ref_60","first-page":"5105709","article-title":"Stacked Autoencoder-Based Deep Neural Network for Achieving Gearbox Fault Diagnosis","volume":"2018","author":"Liu","year":"2018","journal-title":"Math. Probl. Eng."},{"key":"ref_61","doi-asserted-by":"crossref","first-page":"104771","DOI":"10.1016\/j.jafrearsci.2022.104771","article-title":"Modeling the atmospheric dispersion of SO2 from Mount Nyiragongo","volume":"197","author":"Opio","year":"2022","journal-title":"J. Afr. Earth Sci."}],"container-title":["Nitrogen"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2504-3129\/6\/4\/101\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,11,10]],"date-time":"2025-11-10T09:48:37Z","timestamp":1762768117000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2504-3129\/6\/4\/101"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,11,10]]},"references-count":61,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2025,12]]}},"alternative-id":["nitrogen6040101"],"URL":"https:\/\/doi.org\/10.3390\/nitrogen6040101","relation":{},"ISSN":["2504-3129"],"issn-type":[{"value":"2504-3129","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,11,10]]}}}