Measuring poverty in India with machine learning and remote sensing

TL;DR

This paper employs deep learning on census and survey data to estimate living conditions across India, providing a versatile approach to poverty measurement.

Contribution

It introduces a machine learning method that utilizes remote sensing and census data to assess poverty, achieving broad applicability across various outcomes.

Findings

Comparable accuracy to existing methods

Effective across multiple living condition indicators

Demonstrates potential for large-scale poverty assessment

Abstract

In this paper, we use deep learning to estimate living conditions in India. We use both census and surveys to train the models. Our procedure achieves comparable results to those found in the literature, but for a wide range of outcomes.