9004 Problems of information security. Computer systems Проблемы информационной безопасности. Компьютерные системы 2071-8217 14 IoT devices analysis using neural networks ensemble trained on unbalanced sample Применение ансамбля обученных на несбалансированных выборках нейросетей при анализе состояния устройств Интернета вещей 0000-0003-1798-8257 Sukhoparov Mikhail mail@sukhoparovm.ru 0000-0001-6753-2181 Lebedev Ilya isl_box@mail.ru Russian State Hydrometeorological University Saint Petersburg Federal Research Center of Russian Science Academy 03 06 2021 2 127 134

An approach to identifying anomalous situations in network segments of the Internet of Things based on an ensemble of classifiers is considered. Classifying algorithms are tuned for different types of events and anomalies using training samples of different composition. The use of an ensemble of algorithms makes it possible to increase the accuracy of the results due to collective voting. The experiment performed using three neural networks identical in architecture is described. The results of the assessment were obtained both for each classifier separately and with the use of an ensemble

 ensemble of classifiers anomaly detection parasitic traffic information security