Year-Independent Prediction of Food Insecurity Using Classical and Neural Network Machine Learning Methods

Abstract

Current food crisis predictions are developed by the Famine Early Warning System Network, but they fail to classify the majority of food crisis outbreaks with model metrics of recall (0.23), precision (0.42), and f1 (0.30). In this work, using a World Bank dataset, classical and neural network (NN) machine learning algorithms were developed to predict food crises in 21 countries. The best classical logistic regression algorithm achieved a high level of significance (p < 0.001) and precision (0.75) but was deficient in recall (0.20) and f1 (0.32). Of particular interest, the classical algorithm indicated that the vegetation index and the food price index were both positively correlated with food crises. A novel method for performing an iterative multidimensional hyperparameter search is presented, which resulted in significantly improved performance when applied to this dataset. Four iterations were conducted, which resulted in excellent 0.96 for metrics of precision, recall, and f1. Due to this strong performance, the food crisis year was removed from the dataset to prevent immediate extrapolation when used on future data, and the modeling process was repeated. The best “no year” model metrics remained strong, achieving ≥0.92 for recall, precision, and f1 while meeting a 10% f1 overfitting threshold on the test (0.84) and holdout (0.83) datasets. The year-agnostic neural network model represents a novel approach to classify food crises and outperforms current food crisis prediction efforts.

Food price index prediction using time series models: A study of Cereals, Millets and Pulses

“…Additionally, the research emphasized the signi cance of including exogenous variables, such as weather patterns and market trends, in the models to achieve more accurate predictions. Cade Christensen et al [9] For the study, logistic regression and neural network models were constructed and assessed to determine their capacity to predict food crises with high accuracy, while simultaneously minimizing false positives and false negatives. The three metrics namely RMSE, MAE and SMAPE are considered to compare the performances of each model on each of the datasets separately.…”

“…( 5 Food prediction for household staples using time series forecasting techniques such as SARIMA, ETS, and FB Prophet has been a topic of interest for researchers in recent years. One existing work that is relevant to the project is the paper titled "Forecasting food prices and availability in developing countries: An analysis of time series models for maize in Kenya" by Barrett et al (2017) [9].…”

Machine learning techniques for the identification of risk factors associated with food insecurity among adults in Arab countries during the COVID-19 pandemic

“…Data mining and Machine Learning (ML) techniques have been used in several studies as efficient tools for predicting and identifying the risk factors associated with food insecurity [ 20 , 24 – 26 ]. K-Nearest Neighbor (KNN), Random Forest (RF), Logistic regression and Support Vector Machine (SVM) are among the ML models that have been assessed for the prediction of food insecurity [ 24 , 27 , 28 ]. A number of studies utilizing ML techniques have been pivotal by introducing innovative and comprehensive methodologies for data classification and variable identification.…”

Predicting Food Consumption to Reduce the Risk of Food Insecurity in Kazakhstan

“…Nevertheless, forecasting in the context of food security is a relatively new method. Therefore, it is crucial to understand how to forecast the food security crises and suggest effective future steps using the modelling data (Alfred et al 2022;Christensen et al 2021;Lutoslawski et al 2021). Prediction models for total food consumption and each type of food product present a demanding endeavor, particularly due to the multitude of socioeconomic factors influencing them.…”