Abstract. The authors investigate the approach to classification of network condition on the basis of statistical parameters. The shortcomings of the methods of classification of network condition are revealed and basic implementation of the committee of qualifiers is considered. A modification of the committee of qualifiers with use of a neural network as the meta qualifier is proposed. Experiments are carried out for classification of network condition.
Keywords: stacking, classification, machine learning, computing systems, meta-learning.
1 Kharkiv National University of Radioelectronics, Kharkiv, Ukraine,
e-mail: ruban_i@ukr.net.
2
Kharkiv National University of Radioelectronics, Kharkiv, Ukraine,
e-mail: martovytskyi@gmail.com.
3 Taras Shevchenko National University of Kyiv, Kyiv, Ukraine,
e-mail: lukova@ukr.net.