Poverty Prediction

Poverty prediction research aims to develop accurate and efficient methods for estimating poverty levels, primarily to guide resource allocation and policy interventions. Current approaches heavily utilize machine learning, particularly boosting algorithms (like CatBoost and XGBoost) and deep neural networks, often integrating diverse data sources such as household surveys, satellite imagery (including Sentinel-2 data), and temperature readings. These models show promise in improving the speed and accuracy of poverty estimation compared to traditional survey methods, although research is ongoing to optimize data collection strategies and enhance model robustness across diverse geographical contexts.

Papers

November 5, 2024

Analyzing Poverty through Intra-Annual Time-Series: A Wavelet Transform Approach
Mohammad Kakooei, Klaudia Solska, Adel Daoud
Time Series Constructive Approach Discrete Wavelet Transform Multispectral Data Poverty Level Poverty Reduction Poverty Prediction Multidimensional Poverty

May 28, 2024

Use of Boosting Algorithms in Household-Level Poverty Measurement: A Machine Learning Approach to Predict and Classify Household Wealth Quintiles in the Philippines
Erika Lynet Salvador
Machine Learning Model Machine Learning Approach Gradient Boosting Boosting Algorithm Extreme Gradient Boosting Poverty Level Adaptive Boosting Wealth Distribution Poverty Prediction

July 21, 2023

Poverty rate prediction using multi-modal survey and earth observation data
Simone Fobi, Manuel Cardona, Elliott Collins, Caleb Robinson, Anthony Ortiz, Tina Sederholm, Rahul Dodhia, Juan Lavista Ferres
Remote Sensing Image Satellite Imagery Earth Observation Poverty Level Poverty Prediction

February 28, 2023

Novel Machine Learning Approach for Predicting Poverty using Temperature and Remote Sensing Data in Ethiopia
Om Shah, Krti Tallam
Machine Learning Remote Sensing Temperature Measurement Poverty Level Poverty Prediction Poverty Map

November 16, 2022

Can Strategic Data Collection Improve the Performance of Poverty Prediction Models?
Satej Soman, Emily Aiken, Esther Rolf, Joshua Blumenstock
System Performance Learning Based Data Collection Adaptive Sampling Uniform Sampling Random Sampling Poverty Prediction

November 30, 2021

Predicting Poverty Level from Satellite Imagery using Deep Neural Networks
Varun Chitturi, Zaid Nabulsi
Deep Neural Network Satellite Imagery Overhead Imagery Poverty Level Poverty Prediction

Poverty Prediction

Papers

Analyzing Poverty through Intra-Annual Time-Series: A Wavelet Transform Approach

Use of Boosting Algorithms in Household-Level Poverty Measurement: A Machine Learning Approach to Predict and Classify Household Wealth Quintiles in the Philippines

Poverty rate prediction using multi-modal survey and earth observation data

Novel Machine Learning Approach for Predicting Poverty using Temperature and Remote Sensing Data in Ethiopia

Can Strategic Data Collection Improve the Performance of Poverty Prediction Models?

Predicting Poverty Level from Satellite Imagery using Deep Neural Networks