PM2.5 Concentration Prediction Based on Markov Blanke Feature Selection and Hybrid Kernel Support Vector Regression Optimized by Particle Swarm Optimization
Abstract This study employed air quality and meteorological data as research materials and extracted the optimal feature subset by using the approximate Markov blanket-based normal maximum relevance minimum redundancy (nMRMR) algorithm to serve as the input data of the prediction model. In addition,...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Springer
2021-02-01
|
| Series: | Aerosol and Air Quality Research |
| Subjects: | |
| Online Access: | https://doi.org/10.4209/aaqr.200144 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1825197507892740096 |
|---|---|
| author | Lian-Hua Zhang Ze-Hong Deng Wen-Bo Wang |
| author_facet | Lian-Hua Zhang Ze-Hong Deng Wen-Bo Wang |
| author_sort | Lian-Hua Zhang |
| collection | DOAJ |
| description | Abstract This study employed air quality and meteorological data as research materials and extracted the optimal feature subset by using the approximate Markov blanket-based normal maximum relevance minimum redundancy (nMRMR) algorithm to serve as the input data of the prediction model. In addition, a hybrid kernel (HK) was created to improve upon the traditional support vector regression (SVR) model. Particle swarm optimization (PSO) was used to calculate the optimal parameters of hybrid kernel (HK) SVR, which were then used to establish the nMRMR-PSO-HK-SVR model for PM2.5 concentration prediction. The 2016–2019 year air quality and weather data of Wuhan and Tianjin were employed to test the proposed method. The experimental results show that the mean absolute error (MAE), mean absolute percentage error (MAPE), root mean square error (RMSE) and Theil’s inequality coefficient (TIC) of nMRMR-PSO-HK-SVR model are lower than those of SVR, PSO-SVR, nMRMR-SVR and PSO-HK-SVR model. But also, the proposed model could more precisely track moments of sudden PM2.5 concentration change. Thus, the nMRMR-PSO-HK-SVR model has more satisfactory generalizability and can predict PM2.5 concentration more precisely. |
| format | Article |
| id | doaj-art-6de7adb1833447acb717b90a376b33aa |
| institution | Kabale University |
| issn | 1680-8584 2071-1409 |
| language | English |
| publishDate | 2021-02-01 |
| publisher | Springer |
| record_format | Article |
| series | Aerosol and Air Quality Research |
| spelling | doaj-art-6de7adb1833447acb717b90a376b33aa2025-02-09T12:19:51ZengSpringerAerosol and Air Quality Research1680-85842071-14092021-02-0121611810.4209/aaqr.200144PM2.5 Concentration Prediction Based on Markov Blanke Feature Selection and Hybrid Kernel Support Vector Regression Optimized by Particle Swarm OptimizationLian-Hua Zhang0Ze-Hong Deng1Wen-Bo Wang2School of Literature, Law and Economics, Wuhan University of Science and TechnologySchool of Literature, Law and Economics, Wuhan University of Science and TechnologyHubei Province Key Laboratory of Systems Science in Metallurgical Process, Wuhan University of Science and TechnologyAbstract This study employed air quality and meteorological data as research materials and extracted the optimal feature subset by using the approximate Markov blanket-based normal maximum relevance minimum redundancy (nMRMR) algorithm to serve as the input data of the prediction model. In addition, a hybrid kernel (HK) was created to improve upon the traditional support vector regression (SVR) model. Particle swarm optimization (PSO) was used to calculate the optimal parameters of hybrid kernel (HK) SVR, which were then used to establish the nMRMR-PSO-HK-SVR model for PM2.5 concentration prediction. The 2016–2019 year air quality and weather data of Wuhan and Tianjin were employed to test the proposed method. The experimental results show that the mean absolute error (MAE), mean absolute percentage error (MAPE), root mean square error (RMSE) and Theil’s inequality coefficient (TIC) of nMRMR-PSO-HK-SVR model are lower than those of SVR, PSO-SVR, nMRMR-SVR and PSO-HK-SVR model. But also, the proposed model could more precisely track moments of sudden PM2.5 concentration change. Thus, the nMRMR-PSO-HK-SVR model has more satisfactory generalizability and can predict PM2.5 concentration more precisely.https://doi.org/10.4209/aaqr.200144PM2.5Maximum relevance minimum redundancy (MRMR)Hybrid kernelSupport vector regressionPrediction model |
| spellingShingle | Lian-Hua Zhang Ze-Hong Deng Wen-Bo Wang PM2.5 Concentration Prediction Based on Markov Blanke Feature Selection and Hybrid Kernel Support Vector Regression Optimized by Particle Swarm Optimization Aerosol and Air Quality Research PM2.5 Maximum relevance minimum redundancy (MRMR) Hybrid kernel Support vector regression Prediction model |
| title | PM2.5 Concentration Prediction Based on Markov Blanke Feature Selection and Hybrid Kernel Support Vector Regression Optimized by Particle Swarm Optimization |
| title_full | PM2.5 Concentration Prediction Based on Markov Blanke Feature Selection and Hybrid Kernel Support Vector Regression Optimized by Particle Swarm Optimization |
| title_fullStr | PM2.5 Concentration Prediction Based on Markov Blanke Feature Selection and Hybrid Kernel Support Vector Regression Optimized by Particle Swarm Optimization |
| title_full_unstemmed | PM2.5 Concentration Prediction Based on Markov Blanke Feature Selection and Hybrid Kernel Support Vector Regression Optimized by Particle Swarm Optimization |
| title_short | PM2.5 Concentration Prediction Based on Markov Blanke Feature Selection and Hybrid Kernel Support Vector Regression Optimized by Particle Swarm Optimization |
| title_sort | pm2 5 concentration prediction based on markov blanke feature selection and hybrid kernel support vector regression optimized by particle swarm optimization |
| topic | PM2.5 Maximum relevance minimum redundancy (MRMR) Hybrid kernel Support vector regression Prediction model |
| url | https://doi.org/10.4209/aaqr.200144 |
| work_keys_str_mv | AT lianhuazhang pm25concentrationpredictionbasedonmarkovblankefeatureselectionandhybridkernelsupportvectorregressionoptimizedbyparticleswarmoptimization AT zehongdeng pm25concentrationpredictionbasedonmarkovblankefeatureselectionandhybridkernelsupportvectorregressionoptimizedbyparticleswarmoptimization AT wenbowang pm25concentrationpredictionbasedonmarkovblankefeatureselectionandhybridkernelsupportvectorregressionoptimizedbyparticleswarmoptimization |