Multi-scale spatiotemporal graph convolution network for air quality prediction
- PDF / 1,295,897 Bytes
- 15 Pages / 595.224 x 790.955 pts Page_size
- 36 Downloads / 193 Views
Multi-scale spatiotemporal graph convolution network for air quality prediction Liang Ge1,2
· Kunyan Wu1,2 · Yi Zeng1 · Feng Chang1,2 · Yaqian Wang1,2 · Siyu Li1,2
Accepted: 31 October 2020 © Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract Air pollution is a serious environmental problem that has attracted much attention. Air quality prediction can provide useful information for urban environmental governance decision-making and residents’ daily health control. However, existing research methods have suffered from a weak ability to capture the spatial correlations and fail to model the longterm temporal dependencies of air quality. To overcome these limitations, we propose a multi-scale spatiotemporal graph convolution network (MST-GCN), which consists of a multi-scale block, several spatial-temporal blocks and a fusion block. We first divide the extracted features into several groups based on their domain categories, and represent the spatial correlations across stations as two graphs. Then we combine the grouped features and the constructed graphs in pairs to form a multi-scale block that feeds into spatial-temporal blocks. Each spatial-temporal block contains a graph convolution layer and a temporal convolution layer, which can model the spatial correlations and long-term temporal dependencies. To capture the group interactions, we use a fusion block to fuse multiple groups. Extensive experiments on a real-world dataset demonstrate that our model achieves the highest performance compared with state-of-the-art and baseline models for air quality prediction. Keywords Air quality prediction · Deep learning · Graph convolution network · Temporal convolution network
1 Introduction As the economy and the society develop, the problem of urban air pollution has been receiving increasing attention, especially in developing countries (e.g., China and Brazil). Air pollution consists of a mixture of particulate matter (i.e. P M2.5 and P M10 ) and gaseous species (i.e. NO2 , CO, O3 and SO2 ), which have both acute and chronic effects on human health, especially for children, elderly and people with lung and heart diseases [17]. Many epidemiological studies [29, 30] have also shown that exposure to air polluted environment is closely related to various health effects. Long-term (chronic) exposure may lead to deterioration of the respiratory system, damage to the body’s immune system and increase the risk of cardiovascular disease. Short-term (acute) exposure may also cause acute health problems, such as eye irritation and breathing difficulty [17, 21]. According to the World
Health Organization, nine out of ten people now breathe polluted air, which contributes to the deaths of seven million people every year1 . In addition, fine particulate matter and its derivatives may also cause adverse impacts on the environment, such as poor visibility, global climate change [37], and ecological damage [43]. Cities can be plagued by smog, impacting the daily life of the residents negatively. Therefore, u
Data Loading...
