Application of machine learning methods for automated selection of two-dimensional approximants

Modern challenges in processing complex data require new approaches to modeling. One promising direction is the use of machine learning to automate the selection of approximants. This paper presents an approach to the automated selection of two-dimensional approximants using machine learning methods. The automation of approximant selection aims to improve the accuracy of modeling complex data, reduce computational costs, and provide flexibility of algorithms under various conditions. The proposed methods can be used for approximating nonlinear functions with a large number of data points, particularly in computer modeling and engineering analysis tasks. Practical results of algorithm applications are described, demonstrating that the proposed solutions significantly enhance computational efficiency.