The Big Data group applies deep learning methods to various big data sources, especially high resolution satellite data and street imagery, to characterise urban environmental features and exposures, including housing, the social environment, air quality, noise and the transportation environment (mode, street safety, etc.) and to evaluate how urban land use and service delivery policies could impact health. The group also advances methodologies for the use of big data, such as how best to combine satellite and street level images, transferability of prediction models trained in one city to others, and the use of images to evaluate temporal change.
Machine Learning in Public Health (MLPH) Workshop, 34th Conference on Neural Information Processing Systems (NeurIPS 2020) 2020.
Science of the Total Environment, 701, no. 1, pp. 134896, 2020.
Landscape Architecture Frontiers, 7, no. 2, pp. 22-37, 2019.
Scientific Reports, 9, pp. 6229, 2019.
Urban Forestry & Urban Greening, 38, no. 1, pp. 280-285, 2019.