Air Quality Prediction Based on a Spatiotemporal Attention Mechanism

Abstract

With the rapid development of the Internet of Things and Big Data, smart cities have received increasing attention. Predicting air quality accurately and efficiently is an important part of building a smart city. However, air quality prediction is very challenging because it is affected by many complex factors, such as dynamic spatial correlation between air quality detection sensors, dynamic temporal correlation, and external factors (such as road networks and points of interest). Therefore, this paper proposes a long short-term memory (LSTM) air quality prediction model based on a spatiotemporal attention mechanism (STA-LSTM). The model uses an encoder-decoder structure to model spatiotemporal features. A spatial attention mechanism is introduced in the encoder to capture the relative influence of surrounding sites on the prediction area. A temporal attention mechanism is introduced in the decoder to capture the time dependence of air quality. In addition, for spatial data such as point of interest (POI) and road networks, this paper uses the LINE graph embedding method to obtain a low-dimensional vector representation of spatial data to obtain abundant spatial features. This paper evaluates STA-LSTM on the Beijing dataset, and the root mean square error (RMSE) and R-squared ( R 2 ) indicators are used to compare with six benchmarks. The experimental results show that the model proposed in this paper can achieve better performance than the performances of other benchmarks.

Environmental Affection‐Driven English Tense Analysis: A Healthcare Exercise‐Based Corpus Case Study over Public English Environment

“…The attention probability is calculated using the attention mechanism [ 27 , 28 ]. Attention probability can highlight the importance of a specific word to the whole sentence, and the introduction of attention mechanism considers more contextual temporal associations [ 29 ].…”

Bidirectional convolutional LSTM for the prediction of nitrogen dioxide in the city of Madrid

“…Just et al [ 40 ] applied XGBoost to predict PM 2.5 using satellite-derived aerosol optical depth integrated with recursive feature selection technique. Zou et al [ 41 ] applied spatiotemporal attention based LSTM on the Beijing dataset. Ma et al implemented a Bidirectional LSTM (BLSTM) network with Inverse Distance Weighting to predict PM 2.5 concentration at Guangdong, China [ 42 ].…”

Applications of geospatial big data in the Internet of Things

“…The vehicle data can be used for vehicular crowdsensing, with applications in traffic speed estimation and flow prediction (Wang, Zhang, Shao, et al, 2016). Furthermore, there is a significant presence of geospatial big data in the smart environment domain, in disaster monitoring (Fang et al, 2015), air quality management (Chinnaswamy et al, 2019;Xuyao, Hui, Kexin, Yijin, & Jinhang, 2013;Zou et al, 2021), and water and sewage management (Howell, Rezgui, & Beach, 2018). Other domains were also mentioned, such as logistics (Fernández et al, 2017;Fernández, Suárez, Trujillo, Domínguez, & Santana, 2018;Finogeev et al, 2019;Gupta, Sadana, & Gupta, 2020;Kang et al, 2016;Li et al, 2015;Suárez, Trujillo, Domínguez, & José Miguel Santana, 2015), culture and tourism (Benedusi, Chianese, Marulli, & Piccialli, 2015;Chianese, Marulli, Piccialli, Benedusi, & Jung, 2017;Li, Liao, & Huang, 2020;Mello et al, 2019), and smart water (Howell et al, 2018).…”

“…The first group of papers describes IoT applications containing massive data with spatial aspects. The papers in this group were distributed through many domains listed in RQ1, such as transport and traffic (Cao & Wachowicz, 2019;Coşkun, Çakır, & Anbaroğlu, 2020;Dao, Nguyen, Kiran, & Zettsu, 2020;Duan et al, 2020;Hu et al, 2021), smart cities (Anejionu et al, 2019;Kim et al, 2012;Rathore et al, 2016;Unal et al, 2018;Wen et al, 2016), logistics (Fernández et al, 2018;Finogeev et al, 2019;Li et al, 2015;Suárez et al, 2015) and smart environments (Antonić et al, 2016;Fang et al, 2014Fang et al, , 2015Fang et al, , 2017Terziyski, Tenev, Jeliazkov, Jeliazkova, & Kochev, 2020;Xuyao et al, 2013;Zou et al, 2021).…”