Machine Learning Techniques for the Identification of Risk Factors Associated with Food Insecurity among Adults in Arab countries during the COVID-19 Pandemic
Source check: Source checked. Not independent scientific confirmation.
Finding: In the Corona Cooking Survey analysis of 13,443 adults across Arab countries, 9.3% were classified as food insecure; prevalence was highest in Jordan, Palestine, Lebanon, and Saudi Arabia. On a balanced subset of 4,259, Gradient Boosting and Random Forest achieved reported accuracy of 81.6% and 81.4%; SHAP ranked country and age among key predictors. These are cross-sectional associations, not causal risk factors. The primary v1 describes a 70:20:10 train/test/validation split and 10-fold tuning. This is internal evaluation on the study survey, not external validation; the authors report limited country-specific sample sizes.
Technology: Seven machine-learning algorithms classified Food Consumption Score groups; SHAP feature analysis ranked country, age, financial difficulty, depression, and food access among the most influential predictors.
Publication: Preprint · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
- AI contribution
- Seven machine-learning algorithms classified Food Consumption Score groups; SHAP feature analysis ranked country, age, financial difficulty, depression, and food access among the most influential predictors.
- Test and result
- In the Corona Cooking Survey analysis of 13,443 adults across Arab countries, 9.3% were classified as food insecure; prevalence was highest in Jordan, Palestine, Lebanon, and Saudi Arabia. On a balanced subset of 4,259, Gradient Boosting and Random Forest achieved reported accuracy of 81.6% and 81.4%; SHAP ranked country and age among key predictors. These are cross-sectional associations, not causal risk factors.
- Limitation / still unknown
- The primary v1 describes a 70:20:10 train/test/validation split and 10-fold tuning. This is internal evaluation on the study survey, not external validation; the authors report limited country-specific sample sizes.
Evidence sources and annotation context
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
Evidence sources: https://assets-eu.researchsquare.com/files/rs-2545270/v1/7ffd1cb9-91a4-434f-8a36-7edbdd72a5a9.pdf · https://pmc.ncbi.nlm.nih.gov/articles/PMC10505318/
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: paper. Paper: not linked in this catalog; this does not mean no paper exists.
Review date: 22 Nov 2026. Review scope: The primary v1 full text (posted 2023-03-27) reports a defined regional food-insecurity analysis and AI-derived feature ranking, with explicit internal evaluation. Active and archived catalog scans found no matching DOI, title, or same-result record. The source frames features as associations, not causes, and notes limited per-country sample sizes..
No follow-up recorded; this is not a claim that no follow-up exists. Follow-up review scope and date are not recorded. No source-supported relationships recorded.
Editorial assessment: Impact 3/5 · Validation 2/5 · Novelty 3/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.