Assessing the utility of machine learning for predicting food sufficiency: a case study in Malawi

This study explores the potential of applying machine learning (ML) methods to identify and predict areas at risk of food insufficiency using a parsimonious set of publicly available data sources. We combine household survey data that captures monthly reported food insufficiency with remotely sensed...

Full description

Saved in:
Bibliographic Details
Main Authors: Andrew Tomes, Shahrzad Gholami, Didier Alia, Conor Hennessy, Dafeng Xu, Cecilia Bitz, Rahul Dodhia, Juan Lavista Ferres, C. Leigh Anderson
Format: Article
Language:English
Published: Cambridge University Press 2025-01-01
Series:Data & Policy
Subjects:
Online Access:https://www.cambridge.org/core/product/identifier/S2632324925100138/type/journal_article
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850069293341868032
author Andrew Tomes
Shahrzad Gholami
Didier Alia
Conor Hennessy
Dafeng Xu
Cecilia Bitz
Rahul Dodhia
Juan Lavista Ferres
C. Leigh Anderson
author_facet Andrew Tomes
Shahrzad Gholami
Didier Alia
Conor Hennessy
Dafeng Xu
Cecilia Bitz
Rahul Dodhia
Juan Lavista Ferres
C. Leigh Anderson
author_sort Andrew Tomes
collection DOAJ
description This study explores the potential of applying machine learning (ML) methods to identify and predict areas at risk of food insufficiency using a parsimonious set of publicly available data sources. We combine household survey data that captures monthly reported food insufficiency with remotely sensed measures of factors influencing crop production and maize price observations at the census enumeration area (EA) in Malawi. We consider three machine-learning models of different levels of complexity suitable for tabular data (TabNet, random forests, and LASSO) and classical logistic regression and examine their performance against the historical occurrence of food insufficiency. We find that the models achieve similar accuracy levels with differential performance in terms of precision and recall. The Shapley additive explanation decomposition applied to the models reveals that price information is the leading contributor to model fits. A possible explanation for the accuracy of simple predictors is the high spatiotemporal path dependency in our dataset, as the same areas of the country are repeatedly affected by food crises. Recurrent events suggest that immediate and longer-term responses to food crises, rather than predicting them, may be the bigger challenge, particularly in low-resource settings. Nonetheless, ML methods could be useful in filling important data gaps in food crises prediction, if followed by measures to strengthen food systems affected by climate change. Hence, we discuss the tradeoffs in training these models and their use by policymakers and practitioners.
format Article
id doaj-art-07cf17a4e55a47fb8d463d7fb24fd29a
institution DOAJ
issn 2632-3249
language English
publishDate 2025-01-01
publisher Cambridge University Press
record_format Article
series Data & Policy
spelling doaj-art-07cf17a4e55a47fb8d463d7fb24fd29a2025-08-20T02:47:49ZengCambridge University PressData & Policy2632-32492025-01-01710.1017/dap.2025.10013Assessing the utility of machine learning for predicting food sufficiency: a case study in MalawiAndrew Tomes0https://orcid.org/0009-0003-3099-3656Shahrzad Gholami1Didier Alia2https://orcid.org/0000-0001-9792-4557Conor Hennessy3Dafeng Xu4https://orcid.org/0000-0002-2626-2274Cecilia Bitz5Rahul Dodhia6Juan Lavista Ferres7C. Leigh Anderson8Evans School of Public Policy and Governance, https://ror.org/00cvxb145 University of Washington , Seattle, WA, USAhttps://ror.org/00d0nc645 Microsoft AI for Good LabEvans School of Public Policy and Governance, https://ror.org/00cvxb145 University of Washington , Seattle, WA, USAEvans School of Public Policy and Governance, https://ror.org/00cvxb145 University of Washington , Seattle, WA, USAEvans School of Public Policy and Governance, https://ror.org/00cvxb145 University of Washington , Seattle, WA, USAAtmospheric Sciences Department, https://ror.org/00cvxb145 University of Washington , Seattle, WA, USAhttps://ror.org/00d0nc645 Microsoft AI for Good Labhttps://ror.org/00d0nc645 Microsoft AI for Good LabEvans School of Public Policy and Governance, https://ror.org/00cvxb145 University of Washington , Seattle, WA, USAThis study explores the potential of applying machine learning (ML) methods to identify and predict areas at risk of food insufficiency using a parsimonious set of publicly available data sources. We combine household survey data that captures monthly reported food insufficiency with remotely sensed measures of factors influencing crop production and maize price observations at the census enumeration area (EA) in Malawi. We consider three machine-learning models of different levels of complexity suitable for tabular data (TabNet, random forests, and LASSO) and classical logistic regression and examine their performance against the historical occurrence of food insufficiency. We find that the models achieve similar accuracy levels with differential performance in terms of precision and recall. The Shapley additive explanation decomposition applied to the models reveals that price information is the leading contributor to model fits. A possible explanation for the accuracy of simple predictors is the high spatiotemporal path dependency in our dataset, as the same areas of the country are repeatedly affected by food crises. Recurrent events suggest that immediate and longer-term responses to food crises, rather than predicting them, may be the bigger challenge, particularly in low-resource settings. Nonetheless, ML methods could be useful in filling important data gaps in food crises prediction, if followed by measures to strengthen food systems affected by climate change. Hence, we discuss the tradeoffs in training these models and their use by policymakers and practitioners.https://www.cambridge.org/core/product/identifier/S2632324925100138/type/journal_articlecrop pricesfood insufficiencymachine learningMalawiremote sensing
spellingShingle Andrew Tomes
Shahrzad Gholami
Didier Alia
Conor Hennessy
Dafeng Xu
Cecilia Bitz
Rahul Dodhia
Juan Lavista Ferres
C. Leigh Anderson
Assessing the utility of machine learning for predicting food sufficiency: a case study in Malawi
Data & Policy
crop prices
food insufficiency
machine learning
Malawi
remote sensing
title Assessing the utility of machine learning for predicting food sufficiency: a case study in Malawi
title_full Assessing the utility of machine learning for predicting food sufficiency: a case study in Malawi
title_fullStr Assessing the utility of machine learning for predicting food sufficiency: a case study in Malawi
title_full_unstemmed Assessing the utility of machine learning for predicting food sufficiency: a case study in Malawi
title_short Assessing the utility of machine learning for predicting food sufficiency: a case study in Malawi
title_sort assessing the utility of machine learning for predicting food sufficiency a case study in malawi
topic crop prices
food insufficiency
machine learning
Malawi
remote sensing
url https://www.cambridge.org/core/product/identifier/S2632324925100138/type/journal_article
work_keys_str_mv AT andrewtomes assessingtheutilityofmachinelearningforpredictingfoodsufficiencyacasestudyinmalawi
AT shahrzadgholami assessingtheutilityofmachinelearningforpredictingfoodsufficiencyacasestudyinmalawi
AT didieralia assessingtheutilityofmachinelearningforpredictingfoodsufficiencyacasestudyinmalawi
AT conorhennessy assessingtheutilityofmachinelearningforpredictingfoodsufficiencyacasestudyinmalawi
AT dafengxu assessingtheutilityofmachinelearningforpredictingfoodsufficiencyacasestudyinmalawi
AT ceciliabitz assessingtheutilityofmachinelearningforpredictingfoodsufficiencyacasestudyinmalawi
AT rahuldodhia assessingtheutilityofmachinelearningforpredictingfoodsufficiencyacasestudyinmalawi
AT juanlavistaferres assessingtheutilityofmachinelearningforpredictingfoodsufficiencyacasestudyinmalawi
AT cleighanderson assessingtheutilityofmachinelearningforpredictingfoodsufficiencyacasestudyinmalawi