Combining satellite imagery and machine learning to predict poverty.

Publisher: American Association for the Advancement of Science Country of Publication: United States NLM ID: 0404511 Publication Model: Print Cited Medium: Internet ISSN: 1095-9203 (Electronic) Linking ISSN: 00368075 NLM ISO Abbreviation: Science Subsets: MEDLINE

Publication: : Washington, DC : American Association for the Advancement of Science
Original Publication: New York, N.Y. : [s.n.] 1880-

Income* ; Machine Learning*; Developing Countries/*economics ; Poverty/*economics ; Satellite Imagery/*methods; Humans ; Malawi ; Nigeria ; Rwanda ; Tanzania ; Uganda

Reliable data on economic livelihoods remain scarce in the developing world, hampering efforts to study these outcomes and to design policies that improve them. Here we demonstrate an accurate, inexpensive, and scalable method for estimating consumption expenditure and asset wealth from high-resolution satellite imagery. Using survey and satellite data from five African countries--Nigeria, Tanzania, Uganda, Malawi, and Rwanda--we show how a convolutional neural network can be trained to identify image features that can explain up to 75% of the variation in local-level economic outcomes. Our method, which requires only publicly available data, could transform efforts to track and target poverty in developing countries. It also demonstrates how powerful machine learning techniques can be applied in a setting with limited training data, suggesting broad potential application across many scientific domains.
(Copyright © 2016, American Association for the Advancement of Science.)

Comment in: Science. 2016 Aug 19;353(6301):753-4. (PMID: 27540154)

Date Created: 20160820 Date Completed: 20160825 Latest Revision: 20191210

20221213

10.1126/science.aaf7894

27540167