The use of big data and neural networks in precision agriculture to increase crop yield and the sustainability of agricultural production

The article is dedicated to analyzing the potential applications of big data and neural network technologies in precision agriculture to increase crop yields and the sustainability of agricultural production. Based on a review of current research, key trends and gaps in this area have been identified. An original methodology has been proposed, which includes the collection and integration of diverse data sets (remote sensing data, sensor data, agrochemical soil indicators, etc.), their processing using machine learning algorithms, including convolutional neural networks, and the creation of predictive models. Empirical testing of the methodology on a sample of 120 fields in various agroclimatic conditions demonstrated an increase in yield prediction accuracy by 15–20% compared to traditional approaches. Promising directions for optimizing decision-support systems in precision agriculture based on big data analysis have been identified. The results obtained are significant for the development of sustainable agriculture and enhancing global food security.