Artificial Intelligence in Agriculture: A Systematic Review of Crop Yield Prediction and Optimization

Abstract

Agriculture is key to global food security, and Artificial Intelligence is emerging as a catalyst for its modernization. This article explores how advancements are transforming resource management in agriculture and decision-making for crop yield prediction and agricultural optimization. It is based on a Systematic Literature Review (SLR) of 25 selected studies from an initial set of 776 to identify key algorithms, critical variables, and evaluation metrics. The bibliometric analysis reveals a recent increase in publications on the topic, with a focus on algorithms such as SVM, KNN, and XGBoost. Important variables considered include soil data, climate data, and crop characteristics. Challenges identified include model complexity, data quality, and the need for appropriate pre-processing methods. The ultimate goal is to optimize agricultural practices and improve productivity through accurate predictions.