Machine learning-enhanced evaluation of food security across 169 economies

The assessment of global food security is imperative for sustainable development worldwide. However, limited data availability has impeded a comprehensive evaluation on a global scale. This study addresses this challenge by conducting a thorough food security analysis spanning 169 countries from 2008 to 2020. To overcome data limitations, an imputed database is generated using four machine learning models to estimate missing data. The results uncover significant disparities in food security across regions. Europe, Oceania, and North America exhibit higher levels, while East Asia and South America fall in the middle, and Africa and South Asia experience lower levels. Notably, six African countries have made considerable progress in food security, potentially attributed to support from nations such as France, the United States, China, and Japan. Policy recommendations include humanitarian relief, expanding social safety nets, developing transportation networks, supplying fertilizers in vulnerable areas, and stabilizing food prices through national fiscal policies are also proposed for economies with low food security performance. The study provides an imputed database for 44 indicators from 2008 to 2020, offering insights into food security and potential solutions for governments and international organizations.