Reconstructing secondary data based on air quality, meteorological and traffic data considering spatiotemporal components

Ditsuhi Iskandaryan; Francisco Ramos; Sergio Trilles

doi:10.1016/j.dib.2023.108957

Reconstructing secondary data based on air quality, meteorological and traffic data considering spatiotemporal components

Data Brief. 2023 Feb 8:47:108957. doi: 10.1016/j.dib.2023.108957. eCollection 2023 Apr.

Authors

Ditsuhi Iskandaryan¹, Francisco Ramos¹, Sergio Trilles¹

Affiliation

¹ Institute of New Imaging Technologies (INIT), Universitat Jaume I, Av. Vicente Sos Baynat s/n, Castelló de la Plana 12071, Spain.

Abstract

This paper introduces the reconstructed dataset along with procedures to implement air quality prediction, which consists of air quality, meteorological and traffic data over time, and their monitoring stations and measurement points. Given the fact that those monitoring stations and measurement points are located in different places, it is important to incorporate their time series data into a spatiotemporal dimension. The output can be used as input for various predictive analyses, in particular, we used the reconstructed dataset as input for grid-based (Convolutional Long Short-Term Memory and Bidirectional Convolutional Long Short-Term Memory) and graph-based (Attention Temporal Graph Convolutional Network) machine learning algorithms. The raw dataset is obtained from the Open Data portal of the Madrid City Council.

Keywords: Geospatial analysis; Nitrogen dioxide prediction; Secondary data; Spatiotemporal prediction.