Table 2 Sample of the studies related to \(PM_{2.5}\) forecasting.
From: A hybrid Harris Hawks Optimization with Support Vector Regression for air quality forecasting
Study | Publishing year | Dataset | Period | Goal | Evaluation metric | Method |
|---|---|---|---|---|---|---|
2019 | the Greater London Area Pollution Data | 2016:2017 | \(O_3\), \(PM_{10}\) | MAE, RMSE, AI, \(R^2\) | CEeSNN | |
2019 | Collected by auther | 16 mos | \(PM_{2.5}\) | MAE, \(R^2\) | Kernel based linear regression, Bayesian based | |
2020 | Beijing, Wuhan, Shanghai, Guangzhou, Chengdu | 2014 | \(PM_{2.5}\) | \(R^ 2\) | MLR, GWR, RF, GRNN | |
2020 | China | 2014 | \(PM_{2.5}\) | \(R^2\), RMSE, MPE, RPE | GTWNN | |
2020 | Jinan, Nanjing, Chongqing | 2017:2018 | \(PM_{2.5}\),\(PM_{10}\), \(SO_2\),AQI | MAE, RMSE, MAPE, IA, U1,U2, R | ICEEMDAN-MOHHO-ELM | |
2020 | Central Pollution Control in India | Â | \(PM_{2.5}\) | \(R^2\), MAE, MAPE, RMSE | WANFIS-PSO | |
2020 | CPCB | 2016:2018 | \(PM_{2.5}\) | MSE, Pearson Correlation | ANN, SVM | |
2020 | Beijing, Guangzhou, Shanghai | 2016 | \(PM_{2.5}\), \(PM_{10}\) | MAE, RMSE, MAPE, TIC, VAR,IA, DA | FCFS | |
2020 | Beijing | 2010:2014 | \(PM_{2.5}\) | RMSE | AE-BiLSTM | |
2020 | TAQMN | 2012: 2017 | \(PM_{2.5}\) | RMSE, MSE, MAE, \(R^2\) | Gradient Booting Regression | |
2021 | China (14 cities) | 2014:2018 | \(PM_{2.5}\) | RMSE, MAE | FDN-Learning | |
2021 | Beijing | 2015:2017 | \(PM_{2.5}\) | MAPE, MAE, RMSE | CEEMDAN-DeepTCN | |
2021 | Beijing, Shanghai, Wuhan, Guangzhou | 2019 | \(PM_{2.5}\) | MAE, MAPE, RMSE | GB-ELM-MIMO-ECM, MAdaBoost-ELM-MIMO-ECM, GB-ELM-Recursive-ECM, MAdaBoost-ELM-Recursive-ECM | |
2021 | BJ2014 | 2014:2015 | \(PM_{2.5}\) | MAE, RMSE, SMAPE | SpAttRNN | |
BJ2017 | 2017:2018 | |||||
Meteorological , POI | ||||||
2021 | Beijing | 2014:2015 | \(PM_{2.5}\) | RMSE, MAE, \(R^2\) | ST-CausalConvNet | |
2022 | Bei-Shang-Guang-Shen | 2020 | \(PM_{2.5}\) | MAPE, MAE, RMSE, DA | mRMR, BPNN, ELM, GRNN, BiLSTM, MOWCA | |
2022 | Collected by auther | Â | \(PM_{2.5}\), CO, \(SO_2\), \(O_3\), \(H_{2}S\), \(NO_2\),\(PM_{10}\) | MSE | LSTM | |
2022 | Beijing, Tianjin, Shanghai, Chongqing | 2018:2019 | \(PM_{2.5}\), \(PM_{10}\) | MAPE, MAE, RMSE, NRMSE, \(R^2\) | RLMD-ARIMA- RVMcom-MW | |
2022 | Japanese ministry of environment. | 2014 | \(PM_{2.5}\), \(NO_2\) | RMSE, ME, MAE, \(R^2\) | LightGBM, RF,XGBoost | |
2022 | China’s National Ambient Air Quality Monitoring Network | 2015:2020 | \(PM_{2.5}\) | \(R^2\), MAE, MAPE, RMSE | RF | |
2022 | CNEMC | 2013:2020 | \(PM_{2.5}\) | RMSE, \(R^2\), MAE, PRE | STENN | |
2022 | Utrecht, NL | 2020 | \(PM_{2.5}\) | RMSE, MAE, IA, R, Accuracy, NMB, NMSD | AVGAE, GRF | |
Antwerp, BE | 2021 | \(NO_2\) | ||||
Oakland, US | 2019 | \(NO_2\) | ||||
2023 | Beijing \(PM_{2.5}\) | 2010:2014 | \(PM_{2.5}\) | RMSE, MAE | CRINet | |
Beijing Shunyi-Station | 2013:2017 | \(SO_2\), \(NO_2\), \(PM_{2.5}\), \(PM_{10}\), \(O_3\) | ||||
2023 | Lanzhou | 2020:2022 | \(PM_{2.5}\) | MAPE, RMSE, MAE, SMAPe, Dstat | VMD-ARIMA-CNN-TCN | |
2023 | India | 2015:2020 | \(PM_{2.5}\), \(PM_{10}\), \(NO_2\), \(SO_2\), \(O_3\) | MAPE, \(R^2\) | SVR, XGBOOST | |
2023 | University of California Irvine (UCI) Beijing, China | 2010:2014 | \(PM_{2.5}\) | RMSE, \(R^2\), MAE, PRE | RBOSR | |
2023 | Xi’an | 2018:2020 | \(PM_{2.5}\) | RSE, CORR, RMSE, MAE | STF-Net | |
Beijing | 2013:2017 |
