2025 , Volume 30, № 5, p.66-79

This study addresses the problem of forecasting chaotic processes generated by a monitoring system observing an unstable system. The presence of chaotic components in the observational data significantly limits the effectiveness of traditional statistical methods, necessitating novel and unconventional approaches to forecasting. In this paper, the implementation of the forecasting software-algorithmic framework is based on a multi-expert system for data analysis. Preliminary forecasts are generated using software agents which are functioning as weak classifiers. The final predictive decision is made by a supervisory expert through the combined processing of results from a group of independent software experts. In machine learning terms, this forecasting framework corresponds to a stacking algorithm within ensemble decision-making technology.