The task of sentiment analysis is to identify the emotional attitude of the author of the text to the subject or topic under discussion. The relevance of the task is largely related to the development of social networks, online recommendation systems and other services containing a large number of people's opinions on various topics, in particular, about goods, services, offers, events, etc. It is important for marketers, sociologists, administrators, politicians, managers to know people's opinions. The article proposes a solution to the problem of sentiment analysis by decomposing the recognition procedure into two stages using several neural networks and dividing the analyzed texts into homogeneous subsets. The aim of the work is to create a more reliable procedure for determining people's opinions based on their reaction to various messages from the Internet. A decomposition technique for organizing a two-stage process of analyzing the sentiment of texts in Russian by training separate neural networks for each subset of data has been developed and practically implemented. This technique combines two levels of information processing: the first level of a neural network classifier and the second level, which includes several neural network analyzers. The proposed two-stage procedure for analyzing the sentiment of the text makes it possible to ensure the scalability of applications, the independence of neural network settings management and to increase the reliability of estimates.

This study presents an analysis of autoencoder models for the problems of detecting anomalies in network traffic. Results of the training were assessed using open source software on the UNB ICS IDS 2017 dataset. As deep learning models, we considered standard and variational autoencoder, Deep SSAD approaches for a normal autoencoder (AE-SAD) and a variational autoencoder (VAE-SAD). The constructed deep learning models demonstrated different indicators of anomaly detection accuracy; the best result in terms of the AUC metric of 98% was achieved with VAE-SAD model. In the future, it is planned to continue the analysis of the characteristics of neural network models in cybersecurity problems. One of directions is to study the influence of structure of network traffic on the performance indicators of using deep learning models. Based on the results, it is planned to develop an approach of robust identification of security events based on deep learning methods.