Targeting Development Aid with Machine Learning and Mobile Phone Data: Evidence from an Anti-Poverty Intervention in Afghanistan

Citation:

E. Aiken, G. Bedoya, A. Coville, and J. E. Blumenstock, “Targeting Development Aid with Machine Learning and Mobile Phone Data: Evidence from an Anti-Poverty Intervention in Afghanistan,” ACM SIGCAS Computing and Sustainable Societies (COMPASS '20), 2021.
Targeting Development Aid with Machine Learning and Mobile Phone Data: Evidence from an Anti-Poverty Intervention in Afghanistan3.68 MB

Abstract:

Can mobile phone data improve program targeting? By combining rich survey data from a “big push” anti-poverty program in Afghanistan with detailed mobile phone logs from program beneficiaries, we study the extent to which machine learning methods can accurately differentiate ultra-poor households eligible for program benefits from ineligible households. We show that supervised learning methods leveraging mobile phone data can identify ultra-poor households nearly as accurately as survey-based measures of consumption and wealth; and that combining survey-based measures with mobile phone data produces classifications more accurate than those based on a single data source.

Publisher's Version

See also: ml-social - Publications and Preprints
Last updated on 10/08/2021