Intelligent food security model to predict the self-sufficiency status of wheat based on supervised classification algorithms

The study presented an intelligent food security (decision support) model to predict the self-sufficiency status of wheat (IFSMPSSW) in Egypt according to food security markers (features) of wheat (FSMW). These markers have the following attributes: region (Reg.), wheat area (WA), yield, wheat production (Prod.), population (Pop.), average per capita of wheat (APCW), other features, and self-sufficiency status of wheat (SSW) as a prediction class. The proposed model utilizes data mining (DM) classification technique and its algorithms such as Naïve Bayes (NB), iterative Dichotomiser 3 (ID3), random forest (RF), and random tree (RT) algorithms to classify and predict the SSW in Egyptian agriculture regions and their governorates. IFSMPSSW aims to support the state of food security of wheat or other crops to close wheat gap and improve the self-sufficiency ratio of wheat (SRW) in Egypt. It supports decision-makers with useful information and recommendations to take appropriate measures and procedures to reduce the wheat insecurity gap in Egypt. These decisions contribute to combating the failure of food supply chains for wheat and food shortage in local and global markets for commerce. Conflicts, natural disasters, high energy prices, or any combination of these affect the global and regional markets and have an effect on the supply chain and the selling price of wheat and other strategic crops. The accuracy of the prediction results for IFSMPSSW by NB, ID3, RF, and RT was the same and reached 92.6%. In 2021, Egypt's self-sufficiency ratio for wheat (SRW) was 48.2% compared to the SRW predicted by the proposed model, which was 69.6%.