Applying 1D Convolutional Neural Networks to Advance Food Security in Support of SDG 2

Ghada Alturif¹Alaa A. El-Bary²Radwa Ahmed Osman Mohamed^2*

• ¹Princess Nourah bint Abdulrahman University, Riyadh, Saudi Arabia
• ²Arab Academy for Science, Technology and Maritime Transport (AASTMT), Alexandria, Egypt

The goal of this study is to predict how well five countries the US, Saudi Arabia, China, Egypt, and Sweden will do in terms of Sustainable Development Goal 2 (SDG 2), particularly the hunger index scores, between 2025 and 2030. Methods: Historical agricultural, nutritional, and socioeconomic data from 2000 to 2022 were analyses and temporal patterns were extracted using a one-dimensional Convolutional Neural Network (1D-CNN). To guarantee precise and believable predictions, the model was trained and verified using historical data.To represent realistic development trajectories towards SDG 2 targets, forecasts were limited to a range of 0 to 100. Results: By identifying minor temporal trends in line with patterns of world development, the 1D-CNN model showed great accuracy in forecasting changes in hunger index scores. The predictions point to possible advancements in the nations under study in terms of lowering hunger and enhancing food security. Conclusions: Policymakers, international organizations, and sustainability advocates may all benefit from the insightful data that the suggested forecasting technique offers. These forecasts encourage more focused initiatives and efficient use of resources, which will eventually speed up efforts to meet SDG 2 (Zero Hunger).