预测对于人们在现实世界中做出决策至关重要。例如，准确的停车可用性预测 能减缓交通拥堵，减少交通事故的发生。有效的空气质量预测能帮助居民合理安 排出行以及政府进行决策和污染控制。随着物联网技术的发展，越来越多的城市 使用传感器来记录各种实时数据，这使得广泛的时空预测研究任务成为可能。现 实世界中的许多数据具有很强的时空关联性，如交通流量数据、停车数据和空气 质量数据。然而，不同于图像、语音等栅格化的数据，这些数据分布在非欧空间， 因此不适合卷积神经网络（Convolutional Neural Networks, CNNs）等传统模型来建 模。为此，本文将这些时空数据的空间信息建模为图，然后基于图卷积网络（Graph Convolutional Networks, GCNs）来处理这些图数据以获得图嵌入，采用包括一维卷 积神经网络（One-dimensional Convolutional Neural Networks, 1D-CNNs）或门控递归 单元（Gated Recurrent Units, GRUs）在内的深度学习模型来处理历史观测数据以进 行时间维度的时间序列预测。然后将图卷积网络与这些时间序列预测的深度学习模 型相结合，以捕捉时空依赖性，进而用于预测未来的时空数据。

       时空预测任务具有多种场景，单一的预测场景难以表明预测的模型的通用性。 为进一步说明本文所提模型的泛化性与通用性，本文选取了两个不同的具有代表性 的预测场景，即空气质量预测和停车占用率预测来验证本文所提出的图卷积网络模 型。另外，由于不同预测任务的不同属性，例如空气质量受风场影响变化更剧烈， 而停车占用率变化更缓慢，本文对停车占用率预测问题和空气质量预测问题分别采 用了不同的图建模方法。对停车占用率预测而言，由于其数值随时间变化较慢且停 车时长呈指数分布，因此本文对其采用静态图进行建模；对于空气质量预测而言， 空气质量受风场的影响而快速变化，因此本文对其额外采用了动态图进行建模。此 外，为了进一步提高预测性能，我们还提出了一个多图多注意力机制框架，多图神 经网络，它结合了多图卷积网络（Multi-graph Convolutional Networks, MGCNs）和多 种注意机制，以利用潜在的先验信息。最后，基于大规模的真实世界数据集，本文 的模型取得了相对于基准模型的最佳性能，本文提出的多图多注意力机制框架也能 进一步提升已有时空图神经网络模型的预测性能。

Prediction is crucial for people to make decisions in the real world. For example, accurate parking availability predictions can alleviate traffic congestion and reduce traffic accidents. Effective air quality predictions can help residents to rationalize their travel and governments to make decisions and control pollution. With the development of Internet of Things (IoT) technology, an increasing number of cities are using sensors to record a variety of realtime data, making it possible to conduct a wide range of spatio-temporal prediction research tasks. Many data in the real world have strong spatio-temporal correlation, such as traffic flow data, parking data, and air quality data. However, Unlike grid-like data, such as images and speech data, these data are distributed in non-Euclidean space, and thus are not suitable for conventional models such as Convolutional Neural Networks (CNNs). To this end, this paper models the spatial information of these spatio-temporal data as graphs, and then processes these graph data based on Graph Convolutional Networks (GCNs) to obtain graph embeddings, employs deep learning models including one-dimension CNNs (1D-CNNs) or Gated Recurrent Units (GRUs) to process historical observations for time series prediction in the time dimension, and then combines the GNNs with these time-series prediction deep learning models to capture the spatio-temporal dependencies, which in turn can be used for predicting the future spatio-temporal data.

There are multiple scenarios in spatio-temporal prediction tasks, and a single prediction scenario is difficult to demonstrate the generality of the predicted model. To further illustrate the generalization and generality of the proposed model, two different representative prediction scenarios, i.e., air quality prediction and parking occupancy prediction, are selected to validate the proposed graph convolutional network model in this paper. In addition, due to the different properties of different prediction tasks, for example, air quality varies more drastically by wind field, while parking occupancy varies more slowly, different graph modeling methods are used for the parking occupancy prediction problem and the air quality prediction problem, respectively. For the parking occupancy prediction, the parking occupancy changes slowly with time and the parking duration obeys an exponential distribution, so we models it with a static graph. For the air quality prediction, the air quality changes rapidly due to the wind field, so we additionally models it with a dynamic graph. Moreover, to further improve the prediction performance, we also propose a multi-graph multi-attention mechanism framework which combines Multi-graph Convolutional Networks (MGNNs) and multiple attention mechanisms to leverage potential prior information. Finally, based on large-scale real-world datasets, our model achieves the best performance compared to the baseline models. The multi-graph multi-attention mechanism framework proposed in this paper can also further improve the prediction performance of existing spatio-temporal graph neural network models.

 

[1] MANVILLE M, SHOUP D. People, Parking and Cities[J]. Journal of Urban Planning and Development, 2005, 131 : 233 – 245.
[2] LIN T, RIVANO H, LE MOU¨EL F. A Survey of Smart Parking Solutions[J]. IEEE Transactions on Intelligent Transportation Systems, 2017, 18(12) : 3229 – 3253.
[3] Bock F, Di Martino S, Origlia A. Smart Parking: Using a Crowd of Taxis to Sense On-Street Parking Space Availability[J]. IEEE Transactions on Intelligent Transportation Systems, 2020, 21(2) : 496 – 508.
[4] TILAHUN S L, DI MARZO SERUGENDO G. Cooperative multiagent system for parking availability prediction based on time varying dynamic Markov chains[J]. Journal of Advanced Transportation, 2017, 2017.
[5] SHAO W, SALIM F D, GU T, et al. Traveling Officer Problem: Managing Car Parking Violations Efficiently Using Sensor Data[J]. IEEE Internet of Things Journal, 2018, 5(2) : 802 – 810.
[6] SHAO W, TAN S, ZHAO S, et al. Incorporating LSTM Auto-Encoders in Optimizations to Solve Parking Officer Patrolling Problem[J]. ACM Transactions on Spatial Algorithms and Systems (TSAS), 2020, 6(3) : 1 – 21
[7] GUO J, HE H, SUN C. ARIMA-Based Road Gradient and Vehicle Velocity Prediction for Hybrid Electric Vehicle Energy Management[J]. IEEE Transactions on Vehicular Technology, 2019, 68(6) : 5309 – 5320.
[8] ZHANG X, ZHAO Z, ZHENG Y, et al. Prediction of Taxi Destinations Using a Novel Data Embedding Method and Ensemble Learning[J]. IEEE Transactions on Intelligent Transportation Systems, 2020, 21(1) : 68 – 78.
[9] LUO X, LI D, YANG Y, et al. Spatiotemporal traffic flow prediction with KNN and LSTM[J]. Journal of Advanced Transportation, 2019, 2019.
[10] DU B, PENG H, WANG S, et al. Deep Irregular Convolutional Residual LSTM for Urban Traffic Passenger Flows Prediction[J]. IEEE Transactions on Intelligent Transportation Systems, 2020, 21(3) : 972 – 985.
[11] SHAO W, ZHANG Y, GUO B, et al. Parking Availability Prediction with Long Short Term Memory Model[G] // LI S. Green, Pervasive, and Cloud Computing. [S.l.] : Springer International Publishing, 2019 : 124 – 137.
[12] VU H T, HUANG C-C. Parking Space Status Inference Upon a Deep CNN and Multi-Task Contrastive Network With Spatial Transform[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2019, 29(4) : 1194 – 1208.
[13] CHANDRA R, BHATTACHARYA U, BERA A, et al. Traphic: Trajectory Prediction in Dense and Heterogeneous Traffic Using Weighted Interactions[C] // Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2019 : 8483 – 8492.
[14] ZHAO L, SONG Y, ZHANG C, et al. T-GCN: A Temporal Graph Convolutional Network for Traffic Prediction[J]. IEEE Transactions on Intelligent Transportation Systems, 2019, 21(9) : 3848 – 3858.
[15] LI Z, XIONG G, CHEN Y, et al. A Hybrid Deep Learning Approach with GCN and LSTM for Traffic Flow Prediction[C] // 2019 IEEE Intelligent Transportation Systems Conference (ITSC). 2019 : 1929 – 1933.
[16] LI Y, YU R, SHAHABI C, et al. Diffusion Convolutional Recurrent Neural Network: Data-Driven Traffic Forecasting[J]. arXiv preprint arXiv:1707.01926, 2018.
[17] YU B, YIN H, ZHU Z. Spatio-Temporal Graph Convolutional Networks: A Deep Learning Framework for Traffic Forecasting[C] // Proceedings of the 27th International Joint Conference on Artificial Intelligence. 2018 : 3634 – 3640.
[18] KIPF T N, WELLING M. Semi-Supervised Classification with Graph Convolutional Networks[J]. arXiv preprint arXiv:1609.02907, 2016.
[19] STRØM-TEJSEN P, ZUKOWSKA D, WARGOCKI P, et al. The Effects of Bedroom Air Quality on Sleep and Next-day Performance[J]. Indoor air, 2016, 26(5) : 679 – 686.
[20] VANDYCK T, KERAMIDAS K, KITOUS A, et al. Air Quality Co-benefits for Human Health and Agriculture Counterbalance Costs to Meet Paris Agreement Pledges[J]. Nature communications, 2018, 9(1) : 1 – 11.
[21] WANG S, HAO J. Air Quality Management in China: Issues, Challenges, and Options[J]. Journal of Environmental Sciences, 2012, 24(1) : 2 – 13.
[22] ZHANG H, WANG S, HAO J, et al. Air Pollution and Control Action in Beijing[J]. Journal of Cleaner Production, 2016, 112 : 1519 – 1527.
[23] LECLERCQ B, KLUZA J, ANTHERIEU S, et al. Air Pollution-derived PM2:5 Impairs Mitochondrial Function in Healthy and Chronic Obstructive Pulmonary Diseased Human Bronchial Epithelial Cells[J]. Environmental pollution, 2018, 243 : 1434 – 1449.
[24] HAMRA G B, GUHA N, COHEN A, et al. Outdoor Particulate Matter Exposure and Lung Cancer: A Systematic Review and Meta-analysis[J]. Environmental health perspectives, 2014.
[25] SUN X, LIU C, WANG Z, et al. Prenatal Exposure to Residential PM2:5 and Anogenital Distance in Infants at Birth: A Birth Cohort Study from Shanghai, China[J]. Environmental Pollution, 2020, 264 : 114684.
[26] LI G, LI L, LIU D, et al. Effect of PM2:5 Pollution on Perinatal Mortality in China[J]. Scientific Reports, 2021, 11(1) : 1 – 12.
[27] JUNG E M, KIM K-N, PARK H, et al. Association Between Prenatal Exposure to PM2:5 and The Increased Risk of Specified Infant Mortality in South Korea[J]. Environment International, 2020, 144 : 105997.
[28] YI X, ZHANG J, WANG Z, et al. Deep Distributed Fusion Network for Air Quality Prediction[C] // Proceedings of the 24th ACM SIGKDD international conference on knowledge discovery & data mining. 2018 : 965 – 973.
[29] WANG W, GUO Y. Air Pollution PM2:5 Data Analysis in Los Angeles Long Beach with Seasonal ARIMA Model[C] // 2009 international conference on energy and environment technology : Vol 3. 2009 : 7 – 10.
[30] REKHI J K, NAGRATH P, JAIN R, et al. Forecasting Air Quality of Delhi Using ARIMA Model[G] // Advances in Data Sciences, Security and Applications. [S.l.] : Springer, 2020 : 315 – 325.
[31] LIU B-C, BINAYKIA A, CHANG P-C, et al. Urban Air Quality Forecasting Based on Multi- Dimensional Collaborative Support Vector Regression (SVR): A Case Study of Beijing-Tianjin- Shijiazhuang[J]. PloS one, 2017, 12(7) : e0179763.
[32] MAO X, SHEN T, FENG X. Prediction of Hourly Ground-level PM2:5 Concentrations 3 Days in Advance Using Neural Networks with Satellite Data in Eastern China[J]. Atmospheric Pollution Research, 2017, 8(6) : 1005 – 1015.
[33] YU R, YANG Y, YANG L, et al. RAQ–A random Forest Approach for Predicting Air Quality in Urban Sensing Systems[J]. Sensors, 2016, 16(1) : 86.
[34] CHALOULAKOU A, GRIVAS G, SPYRELLIS N. Neural Network and Multiple Regression Models for PM10 Prediction in Athens: A Comparative Assessment[J]. Journal of the Air & Waste Management Association, 2003, 53(10) : 1183 – 1190.
[35] DU S, LI T, YANG Y, et al. Deep Air Quality Forecasting Using Hybrid Deep Learning Framework[ J]. IEEE Transactions on Knowledge and Data Engineering, 2021, 33(6) : 2412 – 2424.
[36] CHIANG Y-L, HSIEH C-L, HUANG H-Y, et al. Urban Area PM2:5 Prediction with Machine Methods: An On-Board Monitoring System[C] // 2018 12th International Conference on Sensing Technology (ICST). 2018 : 25 – 30.
[37] PAK U, MA J, RYU U, et al. Deep Learning-Based PM2:5 Prediction Considering the Spatiotemporal Correlations: A Case Study of Beijing, China[J]. Science of The Total Environment, 2020, 699 : 133561.
[38] LIU B, YAN S, LI J, et al. A Spatiotemporal Recurrent Neural Network for Prediction of Atmospheric PM2.5: A Case Study of Beijing[J]. IEEE Transactions on Computational Social Systems, 2021, 8(3) : 578 – 588.
[39] WANG C, ZHU Y, ZANG T, et al. Modeling Inter-station Relationships with Attentive Temporal Graph Convolutional Network for Air Quality Prediction[C] // Proceedings of the 14th ACM International Conference on Web Search and Data Mining. 2021 : 616 – 634.
[40] YU B, YIN H, ZHU Z. Spatio-Temporal Graph Convolutional Networks: A Deep Learning Framework for Traffic Forecasting[C] // Proceedings of the 27th International Joint Conference on Artificial Intelligence. 2018 : 3634 – 3640.
[41] TAI A P, MICKLEY L J, JACOB D J. Correlations between fine particulate matter (PM2:5) and meteorological variables in the United States: Implications for the sensitivity of PM2:5 to climate change[J]. Atmospheric environment, 2010, 44(32) : 3976 – 3984.
[42] WANG C, ZHU Y, ZANG T, et al. Modeling Inter-Station Relationships with Attentive Temporal Graph Convolutional Network for Air Quality Prediction[C] //WSDM ’21 : Proceedings of the 14th ACM International Conference on Web Search and Data Mining. 2021 : 616 – 634.
[43] GENG X, LI Y, WANG L, et al. Spatiotemporal Multi-Graph Convolution Network for Ride- Hailing Demand Forecasting[C] // Proceedings of the AAAI conference on artificial intelligence : Vol 33. 2019 : 3656 – 3663.
[44] WU Z, PAN S, LONG G, et al. Graph WaveNet for Deep Spatial-Temporal Graph Modeling[C] // Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence, IJCAI- 19. 2019 : 1907 – 1913.
[45] Yu F, Guo J, Zhu X, et al. Real time prediction of unoccupied parking space using time series model[C] // 2015 International Conference on Transportation Information and Safety (ICTIS). 2015 : 370 – 374.
[46] LIU Y, SHA M. Research on Prediction of Traffic Flow at Non-detector Intersections Based on Ridge Trace and Fuzzy Linear Regression Analysis[C] // 2009 International Conference on Computational Intelligence and Security : Vol 2. 2009 : 571 – 575.
[47] BOCK F, DI MARTINO S, ORIGLIA A. A 2-step approach to improve data-driven parking availability predictions[C] // Proceedings of the 10th ACM SIGSPATIAL workshop on computational transportation science. 2017 : 13 – 18.
[48] VLAHOGIANNI E I, KEPAPTSOGLOU K, TSETSOS V, et al. A Real-Time Parking Prediction System for Smart Cities[J]. Journal of Intelligent Transportation Systems, 2016, 20(2) : 192 – 204.
[49] 高广银, 丁勇, 姜枫, et al. 基于BP 神经网络的停车诱导泊位预测[J]. 计算机系统应用, 2017, 26(1) : 236 – 239.
[50] ZHENG Y, RAJASEGARAR S, LECKIE C. Parking Availability Prediction for Sensor-Enabled Car Parks in Smart Cities[C] // 2015 IEEE Tenth International Conference on Intelligent Sensors, Sensor Networks and Information Processing (ISSNIP). 2015 : 1 – 6.
[51] ARJONA J, LINARES M P, CASANOVAS J. A Deep Learning Approach to Real-time Parking Availability Prediction for Smart Cities[C] // Proceedings of the Second International Conference on Data Science, E-Learning and Information Systems. 2019 : 1 – 7.
[52] DUN M, XU Z, CHEN Y, et al. Short-Term Air Quality Prediction Based on Fractional Grey Linear Regression and Support Vector Machine[J]. Mathematical Problems in Engineering, 2020, 2020.
[53] HUANG Y, XIANG Y, ZHAO R, et al. Air Quality Prediction Using Improved PSO-BP Neural Network[J]. IEEE Access, 2020, 8 : 99346 – 99353.
[54] WANG S, LI Y, ZHANG J, et al. PM2:5-GNN: A Domain Knowledge Enhanced Graph Neural Network For PM2:5 Forecasting[C] // Proceedings of the 28th International Conference on Advances in Geographic Information Systems. 2020 : 163 – 166.
[55] FAN W, MA Y, LI Q, et al. A Graph Neural Network Framework for Social Recommendations[J]. IEEE Transactions on Knowledge and Data Engineering, 2020 : 1 – 1.
[56] GLIGORIJEVI´C V, RENFREW P D, KOSCIOLEK T, et al. Structure-Based Protein Function Prediction Using Graph Convolutional Networks[J]. Nature communications, 2021, 12(1) : 1 – 14.
[57] YANG S, MA W, PI X, et al. A Deep Learning Approach to Real-Time Parking Occupancy Prediction in Transportation Networks Incorporating Multiple Spatio-Temporal Data Sources[J]. Transportation Research Part C: Emerging Technologies, 2019, 107 : 248 – 265.
[58] ZHAO L, SONG Y, ZHANG C, et al. T-GCN: A Temporal Graph Convolutional Network for Traffic Prediction[J]. IEEE Transactions on Intelligent Transportation Systems, 2019, 21(9) : 3848 – 3858.
[59] DIAO Z,WANG X, ZHANG D, et al. Dynamic Spatial-Temporal Graph Convolutional Neural Networks for Traffic Forecasting[C] // Proceedings of the AAAI conference on artificial intelligence : Vol 33. 2019 : 890 – 897.
[60] PIAN W, WU Y, QU X, et al. Spatial-Temporal Dynamic Graph Attention Networks for Ride- Hailing Demand Prediction[J]. arXiv preprint arXiv:2006.05905, 2020.
[61] LUO W, ZHANG H, YANG X, et al. Dynamic Heterogeneous Graph Neural Network for Real- Time Event Prediction[C] // Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. 2020 : 3213 – 3223.
[62] ZHOU H, ZHANG F, DU Z, et al. Forecasting PM2:5 Using Hybrid Graph Convolution-Based Model Considering DynamicWind-Field to Offer the Benefit of Spatial Interpretability[J]. Environmental Pollution, 2021, 273 : 116473.
[63] VELI ˇ CKOVI´C P, CUCURULL G, CASANOVA A, et al. Graph Attention Networks[J]. International Conference on Learning Representations, 2018.
[64] GUO S, LIN Y, FENG N, et al. Attention Based Spatial-Temporal Graph Convolutional Networks for Traffic Flow Forecasting[J]. Proceedings of the AAAI Conference on Artificial Intelligence,2019, 33 : 922 – 929.
[65] ZHENG C, FAN X, WANG C, et al. GMAN: A Graph Multi-Attention Network for Traffic Prediction[ C] // Proceedings of the AAAI Conference on Artificial Intelligence : Vol 34. 2020 : 1234 – 1241.
[66] HUANG R, HUANG C, LIU Y, et al. LSGCN: Long Short-Term Traffic Prediction with Graph Convolutional Networks[C] // BESSIERE C. Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence, IJCAI-20. [S.l.] : International Joint Conferences on Artificial Intelligence Organization, 2020 : 2355 – 2361.
[67] CHAI D, WANG L, YANG Q. Bike Flow Prediction with Multi-Graph Convolutional Networks[C] // Proceedings of the 26th ACM SIGSPATIAL international conference on advances in geographic information systems. 2018 : 397 – 400.
[68] LV M, HONG Z, CHEN L, et al. Temporal Multi-Graph Convolutional Network for Traffic Flow- Prediction[J]. IEEE Transactions on Intelligent Transportation Systems, 2020 : 1 – 12.
[69] SHAO W, ZHAO S, ZHANG Z, et al. FADACS: A Few-Shot Adversarial Domain Adaptation Architecture for Context-Aware Parking Availability Sensing[C] // 2021 IEEE International Conference on Pervasive Computing and Communications (PerCom). 2021 : 1 – 10.
[70] LIU K S, GAO J, WU X, et al. On-Street Parking Guidance with Real-Time Sensing Data for Smart Cities[C] // 2018 15th Annual IEEE International Conference on Sensing, Communication, and Networking (SECON). 2018 : 1 – 9.
[71] SHAO W, SALIM F D, GU T, et al. Traveling Officer Problem: Managing Car Parking Violations Efficiently Using Sensor Data[J]. IEEE Internet of Things Journal, 2017, 5(2) : 802 – 810.
[72] SHUMAN D I, NARANG S K, FROSSARD P, et al. The Emerging Field of Signal Processing on Graphs: Extending High-dimensional Data Analysis to Networks and Other Irregular Domains[J]. IEEE signal processing magazine, 2013, 30(3) : 83 – 98.
[73] SKARDINGA J, GABRYS B, MUSIAL K. Foundations and Modelling of Dynamic Networks Using Dynamic Graph Neural Networks: A Survey[J]. IEEE Access, 2021.
[74] KIM J-H, ON K-W, LIMW, et al. Hadamard product for low-rank bilinear pooling[J]. arXiv preprint arXiv:1610.04325, 2016.
[75] AFZALI A, RASHID M, AFZALI M, et al. Prediction of air pollutants concentrations from multiple sources using AERMOD coupled with WRF prognostic model[J]. Journal of Cleaner Production, 2017, 166 : 1216 – 1225.
[76] ZHANG X, CAO R, ZHANG Z, et al. Crowd Flow Forecasting with Multi-Graph Neural Networks[ C] // 2020 International Joint Conference on Neural Networks (IJCNN). 2020 : 1 – 7.
[77] VAN ERVEN T, HARREMOS P. R´enyi divergence and Kullback-Leibler divergence[J]. IEEE Transactions on Information Theory, 2014, 60(7) : 3797 – 3820.
[78] NIEPERT M, AHMED M, KUTZKOV K. Learning convolutional Neural Networks for Graphs[C] // International conference on machine learning. 2016 : 2014 – 2023.
[79] BRUNA J, ZAREMBA W, SZLAM A, et al. Spectral Networks and Locally Connected Networks on Graphs[J]. arXiv preprint arXiv:1312.6203, 2013.
[80] DEFFERRARD M, BRESSON X, VANDERGHEYNST P. Convolutional Neural Networks on Graphs with Fast Localized Spectral Filtering[C] // NIPS’16 : Proceedings of the 30th International Conference on Neural Information Processing Systems. [S.l.] : Curran Associates Inc., 2016 : 3844 – 3852.
[81] CHATTERJEE S, ROY S, DAS A K, et al. Secure Biometric-Based Authentication Scheme Using Chebyshev Chaotic Map for Multi-Server Environment[J]. IEEE Transactions on Dependable and Secure Computing, 2016, 15(5) : 824 – 839.
[82] HAN J, MORAGA C. The influence of the sigmoid function parameters on the speed of backpropagation learning[C] // From Natural to Artificial Neural Computation: International Workshop on Artificial Neural Networks Malaga-Torremolinos, Spain, June 7–9, 1995 Proceedings 3. 1995 : 195 – 201.
[83] Wang G, Giannakis G B, Chen J. Learning ReLU Networks on Linearly Separable Data: Algorithm, Optimality, and Generalization[J]. IEEE Transactions on Signal Processing, 2019, 67(9) : 2357 – 2370.
[84] WANG F, CHENG J, LIU W, et al. Additive margin softmax for face verification[J]. IEEE Signal Processing Letters, 2018, 25(7) : 926 – 930.
[85] SU T, LI Z, KAHN R. Relationships between the planetary boundary layer height and surface pollutants derived from lidar observations over China: regional pattern and influencing factors[J]. Atmospheric Chemistry and Physics, 2018, 18(21) : 15921 – 15935.
[86] LUAN T, GUO X, GUO L, et al. Quantifying the relationship between PM 2.5 concentration, visibility and planetary boundary layer height for long-lasting haze and fog–haze mixed events in Beijing[J]. Atmospheric Chemistry and Physics, 2018, 18(1) : 203 – 225.
[87] RENHE Z, LI Q, ZHANG R. Meteorological conditions for the persistent severe fog and haze event over eastern China in January 2013[J]. Science China Earth Sciences, 2014, 57(1) : 26 – 35.
[88] LI X, CHEN X, YUAN X, et al. Characteristics of particulate pollution (PM2:5 and PM10) and their spacescale-dependent relationships with meteorological elements in China[J]. Sustainability, 2017, 9(12) : 2330.
[89] SHAO W, SALIM F D, SONG A, et al. Clustering Big Spatiotemporal-Interval Data[J]. IEEE Transactions on Big Data, 2016, 2(3) : 190 – 203.
[90] QI Y, LI Q, KARIMIAN H, et al. A Hybrid Model for Spatiotemporal Forecasting of PM2:5 Based on Graph Convolutional Neural Network and Long Short-Term Memory[J]. Science of the Total Environment, 2019, 664 : 1 – 10.
[91] Dubey S R, Chakraborty S, Roy S K, et al. DiffGrad: An Optimization Method for Convolutional Neural Networks[J]. IEEE Transactions on Neural Networks and Learning Systems, 2020, 31(11) : 4500 – 4511.
[92] ANON. IBM Global Parking Survey: Drivers Share Worldwide Parking Woes[R]. 2011.