The third issue of the scientific journal "Modelling and Data Analysis" for 2026 has been published. The new issue brings together research in big data analysis, artificial intelligence, numerical modeling, optimization, and modern teaching methods.
A significant part of the issue focuses on modern methods of artificial intelligence and data analysis. The authors explore the use of deep learning to discover hidden patterns in large-scale educational data, propose an architectural approach to constructing metric spaces based on deep neural network representations, and examine the use of graphs to model dependencies among factors involved in information reliability.
Particular attention is paid to research at the intersection of artificial intelligence and cognitive science. One study presents a model for forecasting digital cognitive burnout in students based on multimodal time series and a cross-modal transformer architecture. The issue also examines the YOLO11n model on the Orange Pi5 and Raspberry Pi4B embedded platforms and an intelligent assistant for applying standardized distributed business models.
The issue also includes research on:
-
finding optimal piecewise rectilinear routes with a specified number of turns;
-
embedding physics-informed neural networks into numerical schemes for modeling dynamical systems;
-
numerical modeling of nonlinear systems, including domain-structure annihilation, the Duffing equation, and solving the Dirichlet problem using the random walk method;
-
computer implementation of psychometric methods for measuring the volume of perception and short-term memory;
-
formalizing educational texts and using higher-order algebraic curves in the teaching of higher mathematics.