許多研究指出，細懸浮微粒（PM2.5）不僅對環境與氣候變遷產生影響，亦對人類健康帶來嚴重威脅。因此，發展具有高準確率的空氣品質預報系統，以提供有效的預警與應變資訊，具有重要意義。臺灣目前的空氣品質預報作業主要仰賴CMAQ （Community Multiscale Air Quality model）空氣品質模式，其所需之氣象參數使用WRF（Weather Research and Forecasting）氣象模式的模擬結果，而排放量資料則來自環境部提供之臺灣空氣污染物排放量清冊（Taiwan Emission Data System, TEDS）。儘管此類基於大氣動力以及污染物傳輸與化學轉化過程的數值模式能有效掌握污染物濃度的變化趨勢，但由於模式各項機制參數化所採用的理想性假設與輸入資料本身存在的不確定性，導致數值預報結果仍存在一定誤差。
隨著人工智慧技術的發展，近年來愈來愈多研究開始導入深度學習演算法，以建構預報能力更佳的空氣品質預報模型。本研究採用卷積神經網路（Convolutional Neural Network, CNN）並搭配一維卷積架構用以擷取濃度隨時間變化特徵，建構具高預報能力之PM2.5濃度預報模型。模型整合了環境部空氣品質監測站之每小時PM2.5濃度觀測值、現行空氣品質預報系統（AQF）的預報結果，以及綜觀天氣型態類別（Synoptic Weather Patterns, SWPs），以預測全臺75個監測站未來72小時內的PM2.5濃度變化。模型訓練資料期間涵蓋2019年10月至2021年9月，並以2021年10月至2022年9月的資料進行驗證與測試，以確保模型具備良好的泛化能力與穩定性。
本研究共建構五種CNN模型，分別為：未納入土地利用型態與SWPs的CNN-noLU模型、未納入SWPs資訊的CNN-BASE模型、結合SWPs資訊之CNN-SWP模型、與CNN-SWP使用相同輸入參數，但進一步採用加權損失函數（weighted loss function）處理PM2.5濃度值數據分布不均問題之CNN-SWPW模型，以及與CNN-BASE使用相同輸入參數，但分空品區建構模型。
評估結果顯示，CNN-BASE可將72小時PM2.5濃度預測的平均根均方誤差（Root Mean Square Error, RMSE）由現行AQF的10.48 μg/m³ 明顯降低至6.88 μg/m³。然而，觀測中屬於高濃度事件（PM2.5日均值 ≥ 35.5 μg/m3）僅占整體樣本的3.4%，導致 CNN-BASE於此濃度範圍的預測準確率僅為26.2%。對比CNN-BASE，CNN-noLU與CNN-AQZ皆具有較高的RMSE與相對誤差（Relative Error, RE），說明地理資訊能提升模型的預報能力且分區建構模型並無太大幫助。
考量臺灣中南部高污染事件常 與不利污染物擴散的氣象條件相關，CNN-SWP藉由加入SWP資訊後，有效提升中、高PM2.5濃度的預報表現。而CNN-SWPW透過加權損失函數強化對少數類別樣本的學習能力，使高濃度事件的預測準確率提升至65.7%。
整體而言，本研究展現了CNN模型應用於空氣品質預報上的潛力，所建構之CNN模型可與現行AQF系統整合，對於提升預報準確性與政策決策支援能力具有重要貢獻。

Many studies have demonstrated that fine particulate matter (PM2.5) adversely impacts the environment, exacerbates climate change, and poses a significant risk to human health. Therefore, developing a highly accurate air quality forecasting system has become an urgent and critical task. In Taiwan, the current operational air quality forecasting system (AQF) is based on the Community Multiscale Air Quality (CMAQ) model, which uses emission data from the Taiwan Emission Data System (TEDS) provided by the Ministry of Environment (MOENV) and meteorological parameters simulated from the Weather Research and Forecasting (WRF) model. Although numerical models grounded in atmospheric dynamics and pollutant governing equations can yield reliable long-term trends in pollutant concentrations, forecasting biases remain due to idealized assumptions in model parameterizations and uncertainties from the input data.
With the advancement of artificial intelligence technologies, an increasing number of recent studies have adopted deep learning algorithms to develop air quality forecasting models with superior predictive capabilities. To extract temporal features of pollutants’ concentrations and construct an accurate forecasting model, this study utilized a Convolutional Neural Network (CNN) with a one-dimensional convolutional operation. The constructed model integrates hourly PM2.5 concentration observed from the air quality monitoring stations provided by the MOENV, simulated from the AQF system, and synoptic weather patterns (SWPs) to forecast PM2.5 concentration over the next 72 hours at 75 monitoring sites across Taiwan. The training dataset spans from October 2019 to September 2021, while the data from October 2021 to September 2022 is used for validation and testing to ensure the generalization of the forecasting model. This study developed five CNN models for PM2.5 forecasting: a baseline model excluding meteorological parameters (CNN-BASE), a model without the land use index (CNN-noLU), a model incorporating synoptic weather patterns (CNN-SWP), a model employing a weighted loss function to mitigate the imbalance in PM2.5 concentrations (CNN-SWPW), and a region-specific model using the same input parameters as CNN-BASE but trained separately for each air quality regions (CNN-AQZ).
The evaluation findings indicate that the CNN-BASE model significantly decreased the root mean square error (RMSE) of the 72-hour PM2.5 forecast from 10.48 μg/m³ of AQF to 6.88 μg/m³. Nevertheless, the restricted representation of high pollution events (daily PM2.5 ≥ 35.5 μg/m³), comprising merely 3.4% of the dataset, resulted in a CNN-BASE accuracy of only 26.2% within this concentration range. In comparison with CNN-BASE, both CNN-noLU and CNN-AQZ demonstrate higher RMSE and relative error (RE), suggesting that the inclusion of geographic information improves forecasting accuracy, while constructing region-specific models provides limited benefit.
Since the high pollution events in Taiwan are often associated with meteorological conditions, the CNN-SWP model showed improved forecasting performance for moderate to high PM2.5 concentrations by incorporating SWPs. The CNN-SWPW model further enhanced the accuracy under high concentration scenarios by strengthening the learning of minority samples through a weighted loss function, achieving an accuracy rate of 65.7% in these challenging situations.
Overall, the results demonstrate the potential of CNN models in operational air quality forecasting and their significant contribution to enhancing the accuracy and decision-support capabilities of existing forecasting systems.

Abadi, M., Barham, P., Chen, J., Chen, Z., Davis, A., Dean, J., ... & Zheng, X. (2016). {TensorFlow}: a system for {Large-Scale} machine learning. In 12th USENIX symposium on operating systems design and implementation (OSDI 16) (pp. 265-283).
Albawi S., Mohammed T., & Al-azawi S. (2017). Understanding of a convolutional neural network. In 2017 International Conference on Engineering and Technology (ICET). IEEE, Akdeniz University, Antalya, Turkey, 274–279.
Appel, K. W., Napelenok, S. L., Foley, K. M., Pye, H. O., Hogrefe, C., Luecken, D. J., ... & Young, J. O. (2017). Description and evaluation of the Community Multiscale Air Quality (CMAQ) modeling system version 5.1. Geoscientific model development, 10(4), 1703-1732.
Baek, J., Hu, Y., Odman, M. T., & Russell, A. G. (2011). Modeling secondary organic aerosol in CMAQ using multigenerational oxidation of semi‐volatile organic compounds. Journal of Geophysical Research: Atmospheres, 116(D22).
Berrocal, V. J., Guan, Y., Muyskens, A., Wang, H., Reich, B. J., Mulholland, J. A., & Chang, H. H. (2020). A comparison of statistical and machine learning methods for creating national daily maps of ambient PM2.5 concentration. Atmospheric Environment, 222, 117130.
Byun, D. W., & Ching, J. K. S. (Eds.). (1999). Science algorithms of the EPA Models-3 community multiscale air quality (CMAQ) modeling system (p. 727). Washington, DC: US Environmental Protection Agency, Office of Research and Development.
Byun, D., & Schere, K. L. (2006). Review of the governing equations, computational algorithms, and other components of the Models-3 Community Multiscale Air Quality (CMAQ) modeling system. Applied mechanics reviews, 59(2), 51-77.
Castelli, M., Clemente, F. M., Popovič, A., Silva, S., & Vanneschi, L. (2020). A machine learning approach to predict air quality in California. Complexity, 2020(1), 8049504.
Chang-Hoi, H., Park, I., Oh, H. R., Gim, H. J., Hur, S. K., Kim, J., & Choi, D. R. (2021). Development of a PM2.5 prediction model using a recurrent neural network algorithm for the Seoul metropolitan area, Republic of Korea. Atmospheric Environment, 245, 118021.
Chen, T. F., Chang, K. H., & Tsai, C. Y. (2017). Modeling approach for emissions reduction of primary PM2.5 and secondary PM2.5 precursors to achieve the air quality target. Atmospheric Research, 192, 11-18.
Cheng, F. Y., Chin, S. C., & Liu, T. H. (2012). The role of boundary layer schemes in meteorological and air quality simulations of the Taiwan area. Atmospheric environment, 54, 714-727.
Cheng, F. Y., Hsu, Y. C., Lin, P. L., & Lin, T. H. (2013). Investigation of the effects of different land use and land cover patterns on mesoscale meteorological simulations in the Taiwan area. Journal of applied Meteorology and Climatology, 52(3), 570-587.
Cheng, F. Y., & Hsu, C. H. (2019). Long-term variations in PM2.5 concentrations under changing meteorological conditions in Taiwan. Scientific reports, 9(1), 6635.
Cheng, F. Y., Lin, C. F., Wang, Y. T., Tsai, J. L., Tsuang, B. J., & Lin, C. H. (2019). Impact of effective roughness length on mesoscale meteorological simulations over heterogeneous land surfaces in Taiwan. Atmosphere, 10(12), 805.
Cheng, F. Y., Feng, C. Y., Yang, Z. M., Hsu, C. H., Chan, K. W., Lee, C. Y., & Chang, S. C. (2021). Evaluation of real-time PM2.5 forecasts with the WRF-CMAQ modeling system and weather-pattern-dependent bias-adjusted PM2.5 forecasts in Taiwan. Atmospheric Environment, 244, 117909.
Ching, J., & Byun, D. (1999). Introduction to the Models-3 framework and the Community Multiscale Air Quality model (CMAQ). Science Algorithms of the EPA Models-3 Community Multiscale Air Quality (CMAQ) Modeling System, 20460.
Choi, J. K., Heo, J. B., Ban, S. J., Yi, S. M., & Zoh, K. D. (2012). Chemical characteristics of PM2.5 aerosol in Incheon, Korea. Atmospheric environment, 60, 583-592.
Chollet, F. (2015). Keras. Available online at: https://github.com/keras-team/keras. Accessed 26 June 2020.
Chuang, M. T., Fu, J. S., Jang, C. J., Chan, C. C., Ni, P. C., & Lee, C. T. (2008). Simulation of long-range transport aerosols from the Asian Continent to Taiwan by a Southward Asian high-pressure system. Science of the total environment, 406(1-2), 168-179.
Cohen, A. J., Brauer, M., Burnett, R., Anderson, H. R., Frostad, J., Estep, K., ... & Forouzanfar, M. H. (2017). Estimates and 25-year trends of the global burden of disease attributable to ambient air pollution: an analysis of data from the Global Burden of Diseases Study 2015. The lancet, 389(10082), 1907-1918.
Eder, B., & Yu, S. (2006). A performance evaluation of the 2004 release of Models-3 CMAQ. Atmospheric Environment, 40(26), 4811-4824.
ENVIRON, A. (2014). User’s guide to the comprehensive air quality model with extensions (CAMx).
Eslami, E., Choi, Y., Lops, Y., & Sayeed, A. (2020). A real-time hourly ozone prediction system using deep convolutional neural network. Neural Computing and Applications, 32(13), 8783-8797.
Gao, Z., Do, K., Li, Z., Jiang, X., Maji, K. J., Ivey, C. E., & Russell, A. G. (2024). Predicting PM2.5 levels and exceedance days using machine learning methods. Atmospheric Environment, 323, 120396.
Grell, G. A., Peckham, S. E., Schmitz, R., McKeen, S. A., Frost, G., Skamarock, W. C., & Eder, B. (2005). Fully coupled “online” chemistry within the WRF model. Atmospheric environment, 39(37), 6957-6975.
Guenther, A. B., Jiang, X., Heald, C. L., Sakulyanontvittaya, T., Duhl, T. A., Emmons, L. K., & Wang, X. (2012). The Model of Emissions of Gases and Aerosols from Nature version 2.1 (MEGAN2. 1): an extended and updated framework for modeling biogenic emissions. Geoscientific Model Development, 5(6), 1471-1492.
Hersbach, H., Bell, B., Berrisford, P., Hirahara, S., Horányi, A., Muñoz‐Sabater, J., ... & Thépaut, J. N. (2020). The ERA5 global reanalysis. Quarterly journal of the royal meteorological society, 146(730), 1999-2049.
Hong, S. Y., Noh, Y., & Dudhia, J. (2006). A new vertical diffusion package with an explicit treatment of entrainment processes. Monthly weather review, 134(9), 2318-2341.
Hsu, C. H., & Cheng, F. Y. (2019). Synoptic weather patterns and associated air pollution in Taiwan. Aerosol and Air Quality Research, 19(5), 1139-1151.
Hsu, C. H., Cheng, F. Y., Chang, H. Y., & Lin, N. H. (2019). Implementation of a dynamical NH3 emissions parameterization in CMAQ for improving PM2.5 simulation in Taiwan. Atmospheric Environment, 218, 116923.
Hsu, C. H., Cheng, F. Y., Chen, C. L., Wu, D. H., Chen, T. Y., Liao, K. F., ... & Zhang, Y. T. (2023). A high-resolution inventory of ammonia emissions from agricultural fertilizer application and crop residue in Taiwan. Atmospheric Environment, 309, 119920.
Huang, W. S., Griffith, S. M., Lin, Y. C., Chen, Y. C., Lee, C. T., Chou, C. C. K., ... & Lin, N. H. (2021). Satellite-based emission inventory adjustments improve simulations of long-range transport events. Aerosol and Air Quality Research, 21(10), 210121.
Iacono, M. J., Delamere, J. S., Mlawer, E. J., Shephard, M. W., Clough, S. A., & Collins, W. D. (2008). Radiative forcing by long‐lived greenhouse gases: Calculations with the AER radiative transfer models. Journal of Geophysical Research: Atmospheres, 113(D13).
Kain, J. S. (2004). The Kain–Fritsch convective parameterization: an update. Journal of applied meteorology, 43(1), 170-181.
Kow, P. Y., Wang, Y. S., Zhou, Y., Kao, I. F., Issermann, M., Chang, L. C., & Chang, F. J. (2020). Seamless integration of convolutional and back-propagation neural networks for regional multi-step-ahead PM2.5 forecasting. Journal of Cleaner Production, 261, 121285.
Kow, P. Y., Chang, L. C., Lin, C. Y., Chou, C. C. K., & Chang, F. J. (2022). Deep neural networks for spatiotemporal PM2.5 forecasts based on atmospheric chemical transport model output and monitoring data. Environmental Pollution, 306, 119348.
Lai, H. C., Dai, Y. T., Le, L. P., Pan, B. H., Lai, L. W., & Hsiao, M. C. (2023). Estimation the effect of accumulated long-range transported pollutants during a PM2.5 event in Taiwan. Atmospheric Pollution Research, 14(6), 101758.
Lee, C. G., Yuan, C. S., Chang, J. C., & Yuan, C. (2005). Effects of aerosol species on atmospheric visibility in Kaohsiung city, Taiwan. Journal of the Air & Waste Management Association, 55(7), 1031-1041.
Lee, M., Lin, L., Chen, C. Y., Tsao, Y., Yao, T. H., Fei, M. H., & Fang, S. H. (2020). Forecasting air quality in Taiwan by using machine learning. Scientific reports, 10(1), 4153.
Lee, J. B., Lee, J. B., Koo, Y. S., Kwon, H. Y., Choi, M. H., Park, H. J., & Lee, D. G. (2022). Development of a deep neural network for predicting 6 h average PM2.5 concentrations up to 2 subsequent days using various training data. Geoscientific Model Development, 15(9), 3797-3813.
Li, M., Zhang, Q., Kurokawa, J. I., Woo, J. H., He, K., Lu, Z., ... & Zheng, B. (2017). MIX: a mosaic Asian anthropogenic emission inventory under the international collaboration framework of the MICS-Asia and HTAP. Atmospheric Chemistry and Physics, 17(2), 935-963.
Liu, B. C., Binaykia, A., Chang, P. C., Tiwari, M. K., & Tsao, C. C. (2017). Urban air quality forecasting based on multi-dimensional collaborative Support Vector Regression (SVR): A case study of Beijing-Tianjin-Shijiazhuang. PloS one, 12(7), e0179763.
Lin, Y., Chiang, Y. Y., Pan, F., Stripelis, D., Ambite, J. L., Eckel, S. P., & Habre, R. (2017, November). Mining public datasets for modeling intra-city PM2.5 concentrations at a fine spatial resolution. In Proceedings of the 25th ACM SIGSPATIAL international conference on advances in geographic information systems (pp. 1-10).
Lin, Y., Mago, N., Gao, Y., Li, Y., Chiang, Y. Y., Shahabi, C., & Ambite, J. L. (2018, November). Exploiting spatiotemporal patterns for accurate air quality forecasting using deep learning. In Proceedings of the 26th ACM SIGSPATIAL international conference on advances in geographic information systems (pp. 359-368).
Lu, F., Xu, D., Cheng, Y., Dong, S., Guo, C., Jiang, X., & Zheng, X. (2015). Systematic review and meta-analysis of the adverse health effects of ambient PM2.5 and PM10 pollution in the Chinese population. Environmental research, 136, 196-204.
Lu, H. Y., Lin, S. L., Mwangi, J. K., Wang, L. C., & Lin, H. Y. (2016). Characteristics and source apportionment of atmospheric PM2.5 at a coastal city in southern Taiwan. Aerosol and Air Quality Research, 16(4), 1022-1034.
MacQueen, J. (1967). Some methods for classification and analysis of multivariate observations. In Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability, Volume 1: Statistics (Vol. 5, pp. 281-298). University of California press.
Masud, M., Alhamid, M. F., & Zhang, Y. (2022). A convolutional neural network model using weighted loss function to detect diabetic retinopathy. ACM Transactions on Multimedia Computing, Communications, and Applications (TOMM), 18(1s), 1-16.
Nguyen, P. H., Dao, N. K., & Nguyen, L. S. P. (2024). Development of machine learning and deep learning prediction models for PM2.5 in Ho Chi Minh City, Vietnam. Atmosphere, 15(10), 1163.
O'shea, K., & Nash, R. (2015). An introduction to convolutional neural networks. arXiv preprint arXiv:1511.08458.
Pan, S., Choi, Y., Jeon, W., Roy, A., Westenbarger, D. A., & Kim, H. C. (2017). Impact of high-resolution sea surface temperature, emission spikes and wind on simulated surface ozone in Houston, Texas during a high ozone episode. Atmospheric environment, 152, 362-376.
Pleim, J. E. (2007). A combined local and nonlocal closure model for the atmospheric boundary layer. Part I: Model description and testing. Journal of Applied Meteorology and Climatology, 46(9), 1383-1395.
Pueschel, R. F., Valin, C. V., Castillo, R. C., Kadlecek, J. A., & Ganor, E. (1986). Aerosols in polluted versus nonpolluted air masses: Long-range transport and effects on clouds. Journal of Applied Meteorology and Climatology, 25(12), 1908-1917.
Sayeed, A., Choi, Y., Eslami, E., Lops, Y., Roy, A., & Jung, J. (2020). Using a deep convolutional neural network to predict 2017 ozone concentrations, 24 hours in advance. Neural Networks, 121, 396-408.
Sayeed, A., Lops, Y., Choi, Y., Jung, J., & Salman, A. K. (2021). Bias correcting and extending the PM forecast by CMAQ up to 7 days using deep convolutional neural networks. Atmospheric Environment, 253, 118376.
Sayeed, A., Eslami, E., Lops, Y., & Choi, Y. (2022). CMAQ-CNN: A new-generation of post-processing techniques for chemical transport models using deep neural networks.
Atmospheric Environment, 273, 118961.
Shi, X., Gao, Z., Lausen, L., Wang, H., Yeung, D. Y., Wong, W. K., & Woo, W. C. (2017). Deep learning for precipitation nowcasting: A benchmark and a new model. Advances in neural information processing systems, 30.
Skamarock, W. C., Klemp, J. B., Dudhia, J., Gill, D. O., Barker, D. M., Duda, M. G., ... & Powers, J. G. (2008). A description of the advanced research WRF version 3. NCAR technical note, 475(125), 10-5065.
Tao, W. K., Simpson, J., & McCumber, M. (1989). An ice-water saturation adjustment. Monthly Weather Review, 117(1), 231-235.
Tao, W. K., Wu, D., Lang, S., Chern, J. D., Peters‐Lidard, C., Fridlind, A., & Matsui, T. (2016). High‐resolution NU‐WRF simulations of a deep convective‐precipitation system during MC3E: Further improvements and comparisons between Goddard microphysics schemes and observations. Journal of Geophysical Research: Atmospheres, 121(3), 1278-1305.
Tewari, M., F. Chen, W. Wang, J. Dudhia, M. A. LeMone, K. Mitchell, M. Ek, G. Gayno, J. Wegiel, and R. H. Cuenca. (2004). Implementation and verification of the unified NOAH land surface model in the WRF model. 20th conference on weather analysis and forecasting/16th conference on numerical weather prediction, Seattle, WA, USA (Vol. 14, pp. 11-15).
Thongthammachart, T., Araki, S., Shimadera, H., Eto, S., Matsuo, T., & Kondo, A. (2021). An integrated model combining random forests and WRF/CMAQ model for high accuracy spatiotemporal PM2.5 predictions in the Kansai region of Japan. Atmospheric Environment, 262, 118620.
Tran, H. D., Huang, H. Y., Yu, J. Y., & Wang, S. H. (2023). Forecasting hourly PM2.5 concentration with an optimized LSTM model. Atmospheric Environment, 315, 120161.
Tsai, Y. I., & Cheng, M. T. (1999). Visibility and aerosol chemical compositions near the coastal area in Central Taiwan. Science of the total environment, 231(1), 37-51.
Tsai, Y. T., Zeng, Y. R., & Chang, Y. S. (2018, August). Air pollution forecasting using RNN with LSTM. In 2018 IEEE 16th Intl Conf on Dependable, Autonomic and Secure Computing, 16th Intl Conf on Pervasive Intelligence and Computing, 4th Intl Conf on Big Data Intelligence and Computing and Cyber Science and Technology Congress (DASC/PiCom/DataCom/CyberSciTech) (pp. 1074-1079). IEEE.
Tukey, J. W. (1977). Exploratory data analysis (Vol. 2, pp. 131-160). Reading, MA: Addison-wesley.
Wang-Li, L. (2015). Insights to the formation of secondary inorganic PM2.5: Current knowledge and future needs. International Journal of Agricultural and Biological Engineering, 8(2), 1-13.
Wang, Y. Z., He, H. D., Huang, H. C., Yang, J. M., & Peng, Z. R. (2025). High-resolution spatiotemporal prediction of PM2.5 concentration based on mobile monitoring and deep learning. Environmental Pollution, 364, 125342.
Wu, J. (2017). Introduction to convolutional neural networks. National Key Lab for Novel Software Technology. Nanjing University. China, 5(23), 495.
Yamartino, R. J. (1993). Nonnegative, conserved scalar transport using grid-cell-centered, spectrally constrained Blackman cubics for applications on a variable-thickness mesh. Monthly Weather Review, 121(3), 753-763.
Yang, S. C., Cheng, F. Y., Wang, L. J., Wang, S. H., & Hsu, C. H. (2022). Impact of lidar data assimilation on planetary boundary layer wind and PM2.5 prediction in Taiwan. Atmospheric Environment, 277, 119064.
Yarwood, G., Rao, S., Yocke, M., & Whitten, G. (2005). Updates to the carbon bond chemical mechanism: CB05. Final report to the US EPA, RT-0400675, 8, 13.
Yu, S., Mathur, R., Pleim, J., Wong, D., Gilliam, R., Alapaty, K., ... & Liu, X. (2014). Aerosol indirect effect on the grid-scale clouds in the two-way coupled WRF–CMAQ: model description, development, evaluation and regional analysis. Atmospheric Chemistry and Physics, 14(20), 11247-11285.
Zaini, N. A., Ean, L. W., Ahmed, A. N., & Malek, M. A. (2022). A systematic literature review of deep learning neural network for time series air quality forecasting. Environmental Science and Pollution Research, 29(4), 4958-4990.
Zhou, S., Wang, W., Zhu, L., Qiao, Q., & Kang, Y. (2024). Deep-learning architecture for PM2.5 concentration prediction: A review. Environmental Science and Ecotechnology, 21, 100400.
Zhu, Y. G., Ioannidis, J. P., Li, H., Jones, K. C., & Martin, F. L. (2011). Understanding and harnessing the health effects of rapid urbanization in China.