Combining satellite imagery andmachine learning to predict poverty

Reliable data on economic livelihoods remain scarce in the developing world, hampering effortsto study these outcomes and to design policies that improve them. Here we demonstrate anaccurate, inexpensive, and scalable method for estimating consumption expenditure and assetwealth from high-resolution satellite imagery. Using survey and satellite data from five Africancountries—Nigeria, Tanzania, Uganda, Malawi, and Rwanda—we show how a convolutionalneural network can be trained to identify image features that can explain up to 75% of thevariation in local-level economic outcomes. Our method, which requires only publicly availabledata, could transform efforts to track and target poverty in developing countries. It alsodemonstrates how powerful machine learning techniques can be applied in a setting withlimited training data, suggesting broad potential application across many scientific domains.

https://dorsu.edu.ph/faculty/jscabrera/wp-content/uploads/2022/12/science.aaf7894.pdf