The most important results of scientific research of the laboratory
- effective methods of forming training samples based on the study of self-learning methods are proposed, including methods of labeling unlabeled data and wrapping methods;
- defined classes and possible training datasets for optimizing neural network settings using semi-supervised learning methods;
- a classification based on the number of used classifiers and their types was developed (on the basis of research on wrapping methods and highlighting their advantages).
The obtained results can be used in the construction of artificial intelligence systems in various fields of science and engineering, implemented, in particular, in the medical and military spheres.
