Data and code for Pseudo-space learning for predicting urban heat exposure risk

Source
DataCite
Published
2026
Country
Australia
License
Creative Commons Attribution 4.0 International
Themes
Public Health, Climate & Temperature, Machine Learning & AI

Get the data from DataCite

About this dataset

This repository contains the data, code, model outputs, manuscript figures and tables, and shapefiles used in the study on pseudo-space learning (PSL) for urban heat exposure.The workflow integrates Google Earth Engine (GEE), QGIS-based spatial preprocessing, R-based machine learning, regression kriging, and pseudo-space learning. Raw remote-sensing data, building footprints, census population data, and Australian statistical boundary files are processed to construct model-ready SA2 training data and SA1 prediction data.To reproduce the workflow, users should first run the raw-data scripts in 1_code/1.1_raw_data_download, followed by the preprocessing scripts in 1_code/1.2_data_preprocessing. These steps generate Training.shp and Testing.shp. For model training and prediction, users only need to run 1.3.1_training.R and 1.3.2_prediction.R. Internal scripts are automatically called by...

How the data was collected

The source metadata does not specify a detailed collection methodology. Provenance: this record was harvested from DataCite and published by figshare. Consult the original source for methodological details, temporal coverage, and access constraints.

Browse sustainability datasets on SusanHub