The current stage of development of the world community is characterized by an everincreasing role of the information sphere and is completely dependent on information resources and technologies, their quality and security. Computerization of all aspects of life has become the main reason that a significant part of the elements of social relations cannot be implemented without the use of new IT in various subject areas, and hence without the implementation of a reliable system of integrated security of the developed information automated systems. This article gives the concept of network traffic, considers the classification of network traffic, in which classification by port numbers, deep packet analysis, stochastic packet analysis, and the use of machine learning were identified. Methods for protecting information using trusted technologies were defined, where a general presentation of trust technologies was considered. The main conclusions are drawn on the prospects of using machine learning to classify network traffic in trusted technologies.

This paper proposes the use of hybrid models based on neural networks and fuzzy systems to build intelligent intrusion detection systems based on the theory of fuzzy rules. The presented system will be able to generate rules based on the results using fuzzy logic neurons. To avoid oversaturation and assist in determining the necessary network topology, training models based on extreme learning machine and regularization theory will be used to find the most significant neurons. In this paper, a type of SQL injection cyberattack is considered, which actively exploits errors in systems that communicate with the database via SQL commands, and for this reason is considered a kind of straightforward attack. The fuzzy neural network architecture used in detecting SQL injection attacks is a multi-component structure. The first two layers of the model are considered as a fuzzy inference system capable of extracting knowledge from data and transforming it into fuzzy rules. These rules help build automated systems for detecting SQL injection attacks. The third layer consists of a simple neuron that has an activation function called a leaky ReLU. The first layer consists of fuzzy neurons, the activation functions of which are Gaussian membership functions of fuzzy sets, defined in accordance with the partitioning of the input variables. The technique uses the concept of a simple linear regression model to solve the problem of choosing the best subsets of neurons. To perform model selection, the paper used the widely used least angular regression (LARS) algorithm.

In recent years, the number of cyberattacks has increased significantly. Most enterprises need reliable protection of the intracorporate networks. Intrusion prevention systems allow timely and automatic response to threats of various kinds that cannot be identified by firewalls, anti-viruses and other security systems. Many companies are represented on the market, providing their signatures to implement intrusion prevention systems developed by manufacturers of network equipment or personal security. There is a need to preserve the confidentiality of these rules with the implementation of the possibility of application on devices commercial users. That’s why systems for the distribution of licensed content to consumer devices are being developed. However, it is necessary to ensure a high level of security of such systems, to avoid leaks of classified data provided by third-party vendors.

The idea of smart homes has been around for several decades and has been described by different authors many times since then. However, there are almost always three aspects in the definitions of the last 20 years. First, home devices must be connected, not only to each other, but also to the Internet. Second, an intelligent way to manage the system is needed, such as a central gateway or smart smartphone apps. Finally, there must be some degree of home automation in the system. A hardware and software complex that meets these requirements can be called a “smart home” system. The system of ensuring the security of the "smart home" is now of great practical importance, which should include measures to protect the IT infrastructure, ensuring the personal safety of residents, ensuring their health, the sanitary condition of the premises, as well as the safety of material assets. It follows from this that the problem of the lack of a thorough study of information security threats and the elaboration of protection of the entire software and hardware complex of the "smart home" system is quite urgent. When solving this problem, an analysis of the main types and characteristics of smart home systems was carried out, and their key vulnerabilities were identified. Also, a study of vulnerabilities in the hardware of smart home systems was carried out; A qualitative assessment of the information security risks of a "smart home" has been carried out and protective measures have been developed to reduce them; A prototype of a fragment of the “smart home” security system has been developed and studied. In an experimental study of threats and vulnerabilities of the developed prototype of a fragment of the "smart home" system, the threat of interception of critical information of the system was studied in detail. Based on the results of the development and research of the Security Inspector, conclusions were drawn about the effectiveness of the use of the intrusion detection module.

The article presents approaches to research on the use of knowledge graphs to generate information security audit issues. The significance of the human factor in the audit of information security as an element of socially significant activity is shown. As a solution to the formation of an objective research base, a method of automatically generating questions based on knowledge graphs capable of identifying a set of cartoons that are subject-object relations is proposed. Existing models and a set of dynamically generated metrics were analyzed. In particular, the subjective method of the five-point Makoto Nagao scale and the machine method of the automatic evaluation system based on the N-gram method. Advantages and disadvantages of research methods are analyzed, task setting for further research is presented.