Accounting for the unaccounted (ACCOUNT): A machine-learning based framework for estimating the population of invisible spaces in support of the SDG Slum Indicator


Within ACCOUNT, I will map and estimate the population of invisible spaces, namely those areas omitted in official statistics but inhabited by people living in deprived areas, such as small slums. The population of such spaces is often not accounted for in official statistics, e.g., within the SDG 11.1.1. indicator on slums. According to UN-Habitat, around one billion people live in slums. However, for several countries, estimates are not available, while for other countries reported data do not accurately reflect the population living in deprived areas. The few existing independent surveys on such areas report dramatically higher numbers of deprived populations as compared to official statistics. The main question, to be addressed in ACCOUNT is: how much of the population living in deprived areas is not accounted for in official statistics due to underreporting and the inability of population surveys (e.g., census) to locate invisible spaces such as small and temporary settlements. Reliable and up-to-date data are urgently required, for planning and service provision, humanitarian response or to address dramatic differences in health outcomes (such as the much lower life expectancies in deprived as compared to better-off areas). Remote sensing (RS) imagery can provide consistent spatial data of deprived areas that can account also for invisible spaces. The main scientific innovation of ACCOUNT lies in the development of a transferable framework that combines RS, machine-learning, Voluntary Geographic Information, and local survey data, to arrive at bottom-up population estimations, accounting for invisible spaces. I will build the project on my extensive experiences on mapping deprived areas using RS, my access to a large dataset of many cities and access to a global expert network. This will help informing the neglected question of whether we need to account for 1 billion people or many more, who are living in deprived areas.


Project number


Main applicant

Dr. M. Kuffer

Affiliated with

University of Twente, Faculty of Geo-Information Science and Earth Observation


01/02/2020 to 31/01/2023