Logo do repositório
 
Publicação

Using Machine Learning Methods to Forecast Air Quality

2022-09-01Artigo científico Acesso aberto
Ver registo simples
dc.contributor.authorLei, Thomas M. T.
dc.contributor.authorSiu, Shirley W. I.
dc.contributor.authorMonjardino, Joana
dc.contributor.authorMendes, Luísa
dc.contributor.authorFerreira, Francisco
dc.contributor.institutionCENSE - Centro de Investigação em Ambiente e Sustentabilidade
dc.contributor.institutionDCEA - Departamento de Ciências e Engenharia do Ambiente
dc.contributor.pblMDPI - Multidisciplinary Digital Publishing Institute
dc.date.accessioned2023-03-16T22:37:30Z
dc.date.available2023-03-16T22:37:30Z
dc.date.issued2022-09-01
dc.descriptionFunding Information: This research was funded by Fundação para a Ciência e Tecnologia, I.P., Portugal, grant number UID/AMB/04085/2020, and the APC was funded by CENSE. Funding Information: The work developed was supported by The Macao Meteorological and Geophysical Bureau (SMG). Publisher Copyright: © 2022 by the authors.
dc.description.abstractDespite the levels of air pollution in Macao continuing to improve over recent years, there are still days with high-pollution episodes that cause great health concerns to the local community. Therefore, it is very important to accurately forecast air quality in Macao. Machine learning methods such as random forest (RF), gradient boosting (GB), support vector regression (SVR), and multiple linear regression (MLR) were applied to predict the levels of particulate matter (PM10 and PM2.5) concentrations in Macao. The forecast models were built and trained using the meteorological and air quality data from 2013 to 2018, and the air quality data from 2019 to 2021 were used for validation. Our results show that there is no significant difference between the performance of the four methods in predicting the air quality data for 2019 (before the COVID-19 pandemic) and 2021 (the new normal period). However, RF performed significantly better than the other methods for 2020 (amid the pandemic) with a higher coefficient of determination (R2) and lower RMSE, MAE, and BIAS. The reduced performance of the statistical MLR and other ML models was presumably due to the unprecedented low levels of PM10 and PM2.5 concentrations in 2020. Therefore, this study suggests that RF is the most reliable prediction method for pollutant concentrations, especially in the event of drastic air quality changes due to unexpected circumstances, such as a lockdown caused by a widespread infectious disease.en
dc.description.versionpublishersversion
dc.description.versionpublished
dc.format.extent14
dc.format.extent1768550
dc.identifier.doi10.3390/atmos13091412
dc.identifier.issn2073-4433
dc.identifier.otherPURE: 56087446
dc.identifier.otherPURE UUID: 910027e0-3928-4236-9295-51120ecd2d70
dc.identifier.otherScopus: 85138708178
dc.identifier.otherWOS: 000859358900001
dc.identifier.otherORCID: /0000-0001-5815-2136/work/140247842
dc.identifier.urihttp://hdl.handle.net/10362/150719
dc.identifier.urlhttps://www.scopus.com/pages/publications/85138708178
dc.language.isoeng
dc.peerreviewedyes
dc.subjectair pollution
dc.subjectair quality
dc.subjectair quality forecast
dc.subjectCOVID-19
dc.subjectgradient boosting
dc.subjectmultiple linear regression
dc.subjectrandom forest
dc.subjectsupport vector regression
dc.subjectEnvironmental Science (miscellaneous)
dc.subjectAtmospheric Science
dc.subjectSDG 3 - Good Health and Well-being
dc.titleUsing Machine Learning Methods to Forecast Air Qualityen
dc.title.subtitleA Case Study in Macaoen
dc.typejournal article
degois.publication.issue9
degois.publication.titleAtmosphere
degois.publication.volume13
dspace.entity.typePublication
rcaap.rightsopenAccess

Ficheiros

Principais
A mostrar 1 - 1 de 1
Miniatura
Nome:
Using_Machine_Learning_Methods_to_Forecast_Air_Quality.pdf
Tamanho:
1.69 MB
Formato:
Adobe Portable Document Format
Ver/Abrir

Coleções

FCT: DCEA - Artigos em revista internacional com arbitragem científica