{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,5]],"date-time":"2026-02-05T12:13:08Z","timestamp":1770293588946,"version":"3.49.0"},"reference-count":22,"publisher":"MDPI AG","issue":"5","license":[{"start":{"date-parts":[[2024,4,25]],"date-time":"2024-04-25T00:00:00Z","timestamp":1714003200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"FCT\u2014Foundation for Science and Technology","award":["UIDB\/00326\/2020"],"award-info":[{"award-number":["UIDB\/00326\/2020"]}]},{"name":"FCT\u2014Foundation for Science and Technology","award":["UIDP\/00326\/2020"],"award-info":[{"award-number":["UIDP\/00326\/2020"]}]},{"name":"FCT\u2014Foundation for Science and Technology","award":["RP\/FCA-09\/2023"],"award-info":[{"award-number":["RP\/FCA-09\/2023"]}]},{"name":"Macao Polytechnic University","award":["UIDB\/00326\/2020"],"award-info":[{"award-number":["UIDB\/00326\/2020"]}]},{"name":"Macao Polytechnic University","award":["UIDP\/00326\/2020"],"award-info":[{"award-number":["UIDP\/00326\/2020"]}]},{"name":"Macao Polytechnic University","award":["RP\/FCA-09\/2023"],"award-info":[{"award-number":["RP\/FCA-09\/2023"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Atmosphere"],"abstract":"<jats:p>It is crucial to speed up the training process of multivariate deep learning models for forecasting time series data in a real-time adaptive computing service with automated feature engineering. Multivariate time series decomposition and recombining (MTS-DR) is proposed for this purpose with better accuracy. A proposed MTS-DR model was built to prove that not only the training time is shortened but also the error loss is slightly reduced. A case study is for demonstrating air quality forecasting in sub-tropical urban cities. Since MTS decomposition reduces complexity and makes the features to be explored easier, the speed of deep learning models as well as their accuracy are improved. The experiments show it is easier to train the trend component, and there is no need to train the seasonal component with zero MSE. All forecast results are visualized to show that the total training time has been shortened greatly and that the forecast is ideal for changing trends. The proposed method is also suitable for other time series MTS with seasonal oscillations since it was applied to the datasets of six different kinds of air pollutants individually. Thus, this proposed method has some commonality and could be applied to other datasets with obvious seasonality.<\/jats:p>","DOI":"10.3390\/atmos15050521","type":"journal-article","created":{"date-parts":[[2024,4,26]],"date-time":"2024-04-26T05:13:41Z","timestamp":1714108421000},"page":"521","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":4,"title":["MTS Decomposition and Recombining Significantly Improves Training Efficiency in Deep Learning: A Case Study in Air Quality Prediction over Sub-Tropical Area"],"prefix":"10.3390","volume":"15","author":[{"given":"Benedito Chi Man","family":"Tam","sequence":"first","affiliation":[{"name":"Department of Informatics Engineering, Centre for Informatics and Systems of the University of Coimbra (CISUC), University of Coimbra, 3004-531 Coimbra, Portugal"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8104-7887","authenticated-orcid":false,"given":"Su-Kit","family":"Tang","sequence":"additional","affiliation":[{"name":"Faculty of Applied Sciences, Macao Polytechnic University, Macao SAR, China"}]},{"given":"Alberto","family":"Cardoso","sequence":"additional","affiliation":[{"name":"Department of Informatics Engineering, Centre for Informatics and Systems of the University of Coimbra (CISUC), University of Coimbra, 3004-531 Coimbra, Portugal"}]}],"member":"1968","published-online":{"date-parts":[[2024,4,25]]},"reference":[{"key":"ref_1","unstructured":"Tam, B.C.M., Tang, S.K., and Cardoso, A. (2022, January 19\u201321). Evaluation of ANN Using Air Quality Tracking in Subtropical Medium-Sized Urban City. Proceedings of the 2022 5th International Conference on Pattern Recognition and Artificial Intelligence (PRAI), Chengdu, China."},{"key":"ref_2","unstructured":"Chaudhary, V., Deshbhratar, A., Kumar, V., and Paul, D. (2018, January 20). Time Series Based LSTM Model to Predict Air Pollutant\u2019s Concentration for Prominent Cities in India. Proceedings of the UDM\u201918, London, UK."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"307","DOI":"10.1007\/s11869-014-0266-7","article-title":"Association of Weather and Air Pollution Interactions on Daily Mortality in 12 Canadian Cities","volume":"8","author":"Vanos","year":"2015","journal-title":"Air Qual. Atmos. Health"},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"847","DOI":"10.1007\/s11869-016-0393-4","article-title":"A Wavelet-Based Approach Applied to Suspended Particulate Matter Time Series in Portugal","volume":"9","author":"Cruz","year":"2016","journal-title":"Air Qual. Atmos. Health"},{"key":"ref_5","first-page":"1767","article-title":"Solar Power Time Series Prediction Using Wavelet Analysis","volume":"10","author":"Gaizen","year":"2020","journal-title":"Int. J. Renew. Energy Res."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"012031","DOI":"10.1088\/1755-1315\/294\/1\/012031","article-title":"Improving Forecasting Accuracy of Daily Energy Consumption of Office Building Using Time Series Analysis Based on Wavelet Transform Decomposition","volume":"294","author":"Fang","year":"2019","journal-title":"IOP Conf. Ser. Earth Environ. Sci."},{"key":"ref_7","first-page":"17766","article-title":"Spectral Temporal Graph Neural Network for Multivariate Time-Series Forecasting","volume":"33","author":"Cao","year":"2020","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"118","DOI":"10.1016\/j.atmosenv.2015.02.030","article-title":"Artificial Neural Networks Forecasting of PM2.5 Pollution Using Air Mass Trajectory Based Geographic Model and Wavelet Transformation","volume":"107","author":"Feng","year":"2015","journal-title":"Atmos. Environ."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"111","DOI":"10.1007\/s11869-010-0084-5","article-title":"Prediction of Ozone Concentrations Using Atmospheric Variables","volume":"4","author":"Tsakiri","year":"2011","journal-title":"Air Qual. Atmos. Health"},{"key":"ref_10","unstructured":"Yi, K., Zhang, Q., Cao, L., Wang, S., Long, G., Hu, L., He, H., Niu, Z., Fan, W., and Xiong, H. (2023). A Survey on Deep Learning Based Time Series Analysis with Frequency Transformation. arXiv."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"191","DOI":"10.1007\/s11869-020-00925-4","article-title":"Air Quality Data Series Estimation Based on Machine Learning Approaches for Urban Environments","volume":"14","author":"Rahimpour","year":"2021","journal-title":"Air Qual. Atmos. Health"},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"857","DOI":"10.1007\/s11869-023-01317-0","article-title":"An Optimized Hybrid Deep Learning Model for PM2.5 and O3 Concentration Prediction","volume":"16","author":"Hu","year":"2023","journal-title":"Air Qual. Atmos. Health"},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"136180","DOI":"10.1016\/j.chemosphere.2022.136180","article-title":"Air Quality Index Forecast in Beijing Based on CNN-LSTM Multi-Model","volume":"308","author":"Zhang","year":"2022","journal-title":"Chemosphere"},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"101291","DOI":"10.1016\/j.uclim.2022.101291","article-title":"Prediction of Air Pollutants for Air Quality Using Deep Learning Methods in a Metropolitan City","volume":"46","author":"Das","year":"2022","journal-title":"Urban Clim."},{"key":"ref_15","unstructured":"(2024, March 20). ECMWF (European Centre for Medium-Range Weather Forecasts) Climate Data Store. Available online: https:\/\/cds.climate.copernicus.eu."},{"key":"ref_16","unstructured":"(2024, March 20). SMG (Direc\u00e7\u00e3o dos Servi\u00e7os Meteorol\u00f3gicos e Geof\u00edsicos de Macau) Concentration of Pollutants, Available online: https:\/\/www.smg.gov.mo\/en\/subpage\/181\/airconcentration."},{"key":"ref_17","unstructured":"(2024, March 20). SMG (Direc\u00e7\u00e3o dos Servi\u00e7os Meteorol\u00f3gicos e Geof\u00edsicos de Macau) Present Weather, Available online: https:\/\/www.smg.gov.mo\/en\/subpage\/73\/actualWeather."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"114332","DOI":"10.1016\/j.eswa.2020.114332","article-title":"Stock Price Prediction Using Deep Learning and Frequency Decomposition","volume":"169","author":"Rezaei","year":"2021","journal-title":"Expert Syst. Appl."},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"124627","DOI":"10.1016\/j.jhydrol.2020.124627","article-title":"Monthly Runoff Time Series Prediction by Variational Mode Decomposition and Support Vector Machine Based on Quantum-Behaved Particle Swarm Optimization","volume":"583","author":"Feng","year":"2020","journal-title":"J. Hydrol."},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Seabold, S., and Perktold, J. (2010, January 28\u201330). Statsmodels: Econometric and Statistical Modeling with Python. Proceedings of the 9th Python in Science Conference (SciPy 2010), Austin, TX, USA.","DOI":"10.25080\/Majora-92bf1922-011"},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"102741","DOI":"10.1016\/j.dsp.2020.102741","article-title":"An Improved Deep Learning Model for Predicting Stock Market Price Time Series","volume":"102","author":"Liu","year":"2020","journal-title":"Digit. Signal Process. A Rev. J."},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Ian, V.K., Tse, R., Tang, S.K., and Pau, G. (2023, January 27\u201329). Transforming from Mathematical Model to ML Model for Meteorology in Macao\u2019s Smart City Planning. Proceedings of the 2023 7th International Conference on E-Commerce, E-Business and E-Government, Online.","DOI":"10.1145\/3599609.3599620"}],"container-title":["Atmosphere"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2073-4433\/15\/5\/521\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,10]],"date-time":"2025-10-10T14:33:46Z","timestamp":1760106826000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2073-4433\/15\/5\/521"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,4,25]]},"references-count":22,"journal-issue":{"issue":"5","published-online":{"date-parts":[[2024,5]]}},"alternative-id":["atmos15050521"],"URL":"https:\/\/doi.org\/10.3390\/atmos15050521","relation":{},"ISSN":["2073-4433"],"issn-type":[{"value":"2073-4433","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,4,25]]}}}