Food Security Analysis and Forecasting: A Machine Learning Case Study in Southern Malawi
Cambridge University Press (Data & Policy)
Ref: SRC-001-MWI-002
Accessed: 3/24/2026
Summary
Peer-reviewed study applying supervised ML algorithms (Random Forest, neural networks, classical models) to MIRA sentinel site household survey data in southern Malawi. Best-performing model achieves F1 of 81% and accuracy of 83% using rCSI threshold-16 dichotomisation with 20 SHAP-selected predictor variables. Random Forest outperforms at community level. Location and self-reported welfare are top predictors. Validates sentinel-site forecasting feasibility for anticipatory action.
View Harvard reference
Gholami, S., Knippenberg, E., Campbell, J., Andriantsimba, D., Kamle, A., Parthasarathy, P., Sankar, R., Birge, C. and Lavista Ferres, J.M. (2022) 'Food Security Analysis and Forecasting: A Machine Learning Case Study in Southern Malawi', Data & Policy, 4, e38. doi:10.1017/dap.2022.24.