Identification of socioeconomic factors influencing global food price security using machine learning

Global concern over food prices and security has been exacerbated by the impacts of armed conflicts such as the Russia Ukraine War, pandemic diseases, and climate change. Traditionally, analyzing global food prices and their associations with socioeconomic factors has relied on static linear regression models. However, the complexity of socioeconomic factors and their implications extend beyond simple linear relationships. By incorporating determinants, critical characteristics identification, and comparative model analysis, this study aimed to identify the critical socioeconomic characteristics and multidimensional relationships associated with the underlying factors of food prices and security. Machine learning tools were used to uncover the socioeconomic factors influencing global food prices from 2000 to 2022. A total of 105 key variables from the World Development Indicators and the Food and Agriculture Organization of the United Nations were selected. Machine learning identified four key dimensions of food price security: economic and population metrics, military spending, health spending, and environmental factors. The top 30 determinants were selected for feature extraction using data mining. The efficiency of the support vector regression model allowed for precise prediction making and correlation analysis. Keywords: environment and growth, global economics, price fluctuation, support vector regression