Abstract
Precision agriculture has emerged as an essential approach to address the critical tasks of optimizing resource usage and increasing crop yields in agriculture. This paper presents an information and analytical subsystem specifically developed for precision agriculture, integrating artificial intelligence (AI), unmanned aerial vehicles (UAVs), and meteorological data for crop yield forecasting. The research utilizes a synthetic dataset containing soil and climatic variables, processed by adaptive machine learning models based on Random Forest algorithms. The developed subsystem demonstrated a high classification accuracy of around 90.6% for crop recommendation and moderate regression performance with $\mathbf{R}^{2}$ values of approximately 0.63 for fertilizer recommendations and 0.57 for yield predictions. These outcomes underline the practical value of the system, enabling farmers to make informed decisions regarding crop selection and resource allocation, thereby promoting sustainable agricultural practices and potentially enhancing crop yields.