Работая с нашим сайтом, вы даете свое согласие на использование файлов cookie. Это необходимо для нормального функционирования сайта, показа целевой рекламы и анализа трафика. Статистика использования сайта отправляется в «Яндекс» и «Google»

Понятно

Научный журнал Моделирование, оптимизация и информационные технологииThe scientific journal Modeling, Optimization and Information Technology

Online media
issn 2310-6018

submit
a manuscript

metadata of articles for the last 2 years

Model and method of forming the project team for countering criminal threats

2022. T.10. № 4. id 1239

Zhirnov A.A.

DOI: 10.26102/2310-6018/2022.39.4.004

The article defines the task of forming the project team for countering criminal threats, which can be solved using the methodology of operations research as an optimization assignment problem. The main drawback of using the classical assignment problem to solve this problem is considered – the possibility of optimization by one criterion only. The problem of multi-criteria selection is regarded. Major methods of multi-criteria optimization are listed. Two main groups of these methods are specified. One of them is examined with reference to a set of criteria by their linear additive convolution into super criteria. Some disadvantages of this approach are indicated. Based on this approach, the author formulated a variation of the model and the method for solving the multi-criteria assignment problem with some leveling of the identified shortcomings. The proposed author's approach employs the convolution of criteria by deviating from the ideal point with the measurement of distance in Euclidean space. Possible limitations to the application of the author's version of the method for solving the multi-criteria assignment problem in practice are indicated for which reason a special heuristic method is suggested that helps to level them. The algorithm of the method for forming the project team for countering criminal threats is given.

Keywords: project for countering criminal threats, assignment problem, multi-criteria assignment problem, project team

Algorithm for detecting sources of malicious requests in cyber-physical systems

2022. T.10. № 3. id 1238

Iskhakova A.O. Iskhakov A.Y. Bogacheva D.N. Molotov A.A.

DOI: 10.26102/2310-6018/2022.38.3.020

The paper is devoted to solving the problem of algorithmic security management processes of cyber-physical systems by detecting malicious requests from a number of other associated systems, internal services or human actions. The relevance of the research is due to the high degree of criticality of protection against possible degradation of services as part of the implementation of attacks on compound complex systems responsible for the integration of computing resources into physical entities. The authors focus on denial-of-service attacks on cyber-physical systems by sending http-flood to web management interfaces. The proposed algorithm for detecting malicious requests analyzes the activity of all investigated components of cyber-physical system web services. The research employs the method of visual analysis and data processing based on the representation as a single normalized set. Raw data of the analyzed queries is grouped in a specific way to detect a particular deviation as a suspected threat. Examples of data changes and security system responses are given. Experimental results confirm that the suggested algorithmic software achieves first- and second-order error reduction compared to commonly used regression models in modern application-level firewalls.

Keywords: information security, malicious requests, sources of malicious requests, cyber security, data analysis, threats, denial of service, DDoS, URI, HTTP

Analysis of the possibilities of building a rational structure of the cyber-physical system

2023. T.11. № 1. id 1235

Avetisyan T.V. Lvovich Y.E. Preobrazhenskiy A.P.

DOI: 10.26102/2310-6018/2023.40.1.020

The analysis shows that cyber-physical systems together with cyber-biological and cyber-social systems are now considered as key elements in modern infotelecommunication systems. The concept of a cyber-physical system is based on a dualism of the physical and cybernetic environments. Due to the fact that the physical, biological and social environments are combined and cyberspace is further introduced, there are significant opportunities for the implementation of a wide variety of functions. At the same time, new problems appear, for example, those associated with modeling the processes within such systems. There is also the issue of monitoring within the cyber-physical system since the data can be missed. Therefore, this article aims to develop such a structure of a cyber-physical systems so that its efficiency would be as high as possible. The paper proposes a procedure for the formation of a rational structure of such a system. The components are chosen according to the principle of ranking components relative to their value from the system's point of view. Two types of constraints are used, the first of which are related to the area of the system in question, and the second are related to the technologies employed. At the initial stage, experts implement the choice of components of a cyber-physical system by means of the information system. Next, two selection procedures are applied. As the key result, a structural scheme related to the optimal choice of components of a cyber-physical system is suggested.

Keywords: cyber-physical system, structure, optimization, expert, information system

Methodology for assessing the current state of the engineering telecommunications infrastructure of a special-purpose communications network segment

2022. T.10. № 3. id 1234

Popov A.V. Kanavin S.V. Gilev I.V. Khokhlov N.S. Udalov V.P.

DOI: 10.26102/2310-6018/2022.38.3.023

The article proposes a methodology for assessing the current state of the engineering and telecommunications infrastructure of a special-purpose communication network segment and tested it by the example of the regional segment of the integrated multiservice telecommunications system of the Ministry of Internal Affairs of Russia. A regional segment of a special-purpose communication network is defined as a physical or logical zone in which granting access to resources or the denial of this access are regulated by access rules and control mechanisms. Such zone has a clear boundary with other segments. Taking into consideration the need to maintain state operability of the regional segment of a special-purpose communication network, the task of assessing the current state of the engineering and telecommunications infrastructure is relevant. The paper proposes a sequence of actions aimed at conducting an audit of communication nodes at all levels of the regional segment including engineering infrastructure, telecommunications equipment, data transmission channels. As a mathematical apparatus, mathematical methods for processing expert assessments, associated with determining the significance of individual components of engineering and telecommunications infrastructure, are used. The method of hierarchy analysis with the involvement of expert groups is applied to define the significance coefficients of the factors accounted for when calculating the integral evaluation functions of the regional segment of a special-purpose communication network.

Keywords: audit of the regional segment of a communication network, transmission channels, monitoring of telecommunications equipment, assessment methodology, expert assessments, telecommunications and engineering infrastructure

Optimization of resource support at a given planning horizon of the organizational system development process using visual expert modeling

2022. T.10. № 3. id 1231

Lvovich A.I. Preobrazhenskiy A.P.

DOI: 10.26102/2310-6018/2022.38.3.015

The paper deals with the issues concerning resource support management in the development of an organizational system at a given planning horizon. To solve them, integrating visual expert and optimization modeling within a single algorithmic scheme is proposed. The first task is aimed at determining the importance of the monitored indicators of the organizational system functioning in the implementation of the development process. To address it, a visualization of the initial data is suggested, which helps to accelerate and improve the accuracy of expert assessments when choosing the structure of the forecast model. It is targeted at effective processing of monitoring assessment data by means of visualization techniques The second task is a multi-alternative optimization problem that uses the solutions of the first task and ensures the distribution of the integral amount of resource support to increase the level of the indicators most crucial for development. The third task characterizes management decisions on the distribution of resource support between time ranges at a given planning horizon. It is shown that the combination of visual expert and optimization modeling makes it possible to find the optimal distribution which is more consistent with the real functioning of the organizational system compared to the traditional method of multi-step process of making optimal decisions. The solutions of these problems are combined within an algorithmic scheme and each procedure includes certain actions for data visualization, forecasting, examination and making optimal decisions.

Keywords: organizational system, management, resource support, development, visualization, forecasting, expert assessment, optimization

Development of a matrix for SWOT analysis based on key parameters and criteria taking into account the specifics of managing a medical organization

2022. T.10. № 3. id 1229

Akulova A.D. Korovin E.N.

DOI: 10.26102/2310-6018/2022.38.3.030

The article describes the method of strategic planning – SWOT analysis – which helps to study in detail a set of parameters of both the internal environment of the institution in question and the external environment that has the strongest impact on its functioning whereby management decisions aimed at qualitative and quantitative improvement of performance indicators are made. The features of the management system and operation of a medical organization are analyzed using a key quality indicator – annual statistical reporting reflecting the results of the institution's activities in the main areas for higher state bodies with authority in the field of public health. The analysis of the form composition of accounting and analytical registers and statistical indicators was carried out, and the method of system dynamics was applied in the context of selected management processes to identify the optimal set of parameters of the internal and external environment providing the means for developing a SWOT matrix suitable for the qualitative assessment of the outpatient clinic operation. Criteria, a number of factors and actors that have the greatest impact on the control mechanism by building a tree of goals are identified. The developed SWOT matrix based on a set of key parameters and criteria is a universal tool for strategic planning and can be put to practice when conducting a SWOT analysis of medical organizations various regions.

Keywords: medical organization, SWOT analysis, accounting and analytical registers, goal tree, system dynamics, parameters, internal environment, external environment, statistical reporting

A resource-saving method of distributed computation planning in fog-computing environment

2022. T.10. № 3. id 1228

Klimenko A.B.

DOI: 10.26102/2310-6018/2022.38.3.019

The issues of organizing distributed computation in fog-environments are currently relevant due to the increasing amount of data circulating over global networks. Research carried out in the field of the development of new models, methods and technical means of the fog computing concept covers a wide range of topics, including resource sharing, computational planning, user authentication, and data security. Papers on resource consumption are also presented, specifically those that explore the issue of extending the expedient service life of fog devices, which have a significant impact on the system operating cost. In this article, the solution to the problem of resource saving in this aspect is associated with a reasonable distribution of the computational load over the fog nodes which affects the device indicators, such as the probability of failure-free operation, gamma-percentage time between failures and the average residual resource of a computing device. A method for evaluating the feasibility of placing a computational load on nodes as part of a "greedy" strategy is proposed, as well as a method for selecting nodes to place the load. Combining these methods constitutes a method for distributed computing planning in the fog layer of a network with optimization according to the criterion of resource-saving. The conducted experiment demonstrates the applicability of the developed method and helps to choose the area for further research.

Keywords: resource-saving, computational planning, fog computing,

Authorship identification of a heterogeneous source code for the purposes of cybersecurity management

2022. T.10. № 3. id 1227

Romanov A.S. Kurtukova A.V. Shelupanov A.A. Fedotova A.M.

DOI: 10.26102/2310-6018/2022.38.3.016

The article is devoted to the issue of identifying the author of a heterogeneous source code program by means of a hybrid neural network. The solutions to this problem are especially relevant to the fields of information security, educational process, and copyright protection. The article analyzes modern methods of addressing this problem. The authors propose their own methodology based on a proven in early studies hybrid neural network aimed at evaluating the effectiveness of this approach in simple and difficult cases. This research incorporates experiments on previously unconsidered cases of source code author identification based on heterogeneous data. Cases relevant to corporate development are examined including the analysis of source codes presented as commits and model training on datasets with more than two programming languages. Additionally, the trend of determining the authorship of an artificially generated source code, which is gaining traction, is regarded. A dataset was generated, and an appropriate experiment was performed for each case. The effectiveness of the author's methodology for all three difficult cases was evaluated using a 10 blocks cross-validation. The average accuracy for mixed datasets was 87 % for two programming languages and 76 % for three or more languages, respectively. The average accuracy of the methodology for authorship identification of artificially generated source codes was 81.5 %. Identification of the author of a program source code based on commits was carried out with an accuracy of 84 %. Experiments have shown that the effectiveness of the methodology can be improved in all three cases by using large amounts of training data.

Keywords: authorship, source code, commits, generation, neural network

Vulnerability base formation algorithm and neural network architecture selection for its processing

2022. T.10. № 3. id 1226

Sobolevskaya E.Y. Shevchenko I.D. Alekseev S.E.

DOI: 10.26102/2310-6018/2022.38.3.025

The article discusses the need for an algorithm to form the information system vulnerability base and the selection of the neural network architecture. A description of existing systems and criteria for assessing vulnerabilities as well as a group of metrics are given. The vulnerability databases were analyzed and discrepancies in the assessment of vulnerabilities, advantages and disadvantages were identified. The following architectures were identified and studied: feed forward neural network, generative adversarial network, Autoencoder, recurrent neural network without long short-term memory, recurrent neural network with long short-term memory, Rumelhart multilayer perceptron, liquid state machine, Boltzmann machine. A preliminary analysis of neural network architectures is presented taking into account significant parameters for further use in the field of information security and vulnerability classification. Based on the results obtained during the study of the parameters of neural networks, feed forward neural network, recurrent neural network with long short-term memory and generative adversarial network were identified. An alternative method of forming a vulnerability database by means of neural networks is proposed. As a result, an algorithm for forming a vulnerability base and a method for automating it using a neural network are suggested. The solution will allow the neural network to constantly receive up-to-date data for training and, owing to this, the vulnerability database will be updated as quickly as possible, which will make it the most complete, reliable and up-to-date of all existing vulnerability databases.

Keywords: vulnerabilities, neural networks, neural network architecture, algorithm, threat

The concept of the agent-based model for predicting a patient’s general health in the process of aging

2022. T.10. № 4. id 1225

Lisovenko A.S. Limanovskaya O.V. Gavrilov I.V. Meshchaninov V.N. Myakotnykh V.S.

DOI: 10.26102/2310-6018/2022.39.4.007

Agent-based modeling is actively used for modeling human health. The main advantages of an agent-based approach in this field are the capability to implement a modular approach to health and to account for individual patient indicators. The article presents the concept of a flexible and expandable agent model of the patient, which performs a long-term prediction of the patient's condition based on short-term test treatments administered to them, including geroprophylactic, and by predicting the patient's reaction to exposure in order to prevent future possible diseases with regard to both calendar and biological age. All interactions of the model agents are reduced to assessing the effectiveness of the anti-aging measures in the form of a calculated bio-age which characterizes the degree of decrease in the functional capacity of the organism. As part of the concept, the central agents “Patient”, “Aging Process” and “Impact” are highlighted in the model as well as a number of lower-level agents associated with the agent “Patient”. Lower-level agents are responsible for modeling the physiological processes of body systems or diseases, for example, a chronic disease is allocated its own agent, which affects the patient's condition during the modeling. The types of model agents are extensible, which makes it possible to develop this concept of the model. The paper presents the testing of the agent model concept to identify the effectiveness of the impact on the patient following on from the assessment of changes in the biological age before and after geroprophylactic therapy.

Keywords: agent modeling, patient's health, geroprophylactic treatment, predicting the efficiency of treatment, bioage

Researching and modeling the organizational culture of regional competitive machine-building enterprises

2022. T.10. № 3. id 1224

Rodionova V.O. Fedorkova N.V.

DOI: 10.26102/2310-6018/2022.38.3.008

The paper is devoted to the modeling of an organizational culture by means of an experimental study using a questionnaire and expert a priori ranking. The influence of an organizational culture on the competitiveness of an enterprise is considered. The a priori ranking method makes it possible to objectively assess the subjective opinions of experts and develop a model of factor rank ordering that fully captures the essence of regional machine-building enterprises. For clarity, the results of the study are laid out in the form of a rank histogram. Experts are divided into two groups: managers and workers, employees. In reliance on their opinions, the ranks of social and economic factors are determined. Following on from the study, the factors that fully captures the essence of this enterprise are ascertained. As a result, organizational culture analysis is carried out. In addition to modeling the system of an enterprise organizational culture, a study was conducted with a view to identifying the qualities of managers. As a result, the state of the organizational culture of regional machine-building competitive enterprises was assessed and ways of improving it were determined. A comparative characteristic of the influence of socio-economic factors on the organizational culture of two enterprises in the region has been obtained. Based on the findings that characterize the organizational culture of two competitive machine-building enterprises, the organizational culture was described from the perspective of management and employees in terms of the main parameters. The conclusion is made on the analysis of the diagnostics of the state of the organizational culture and a number of measures are proposed to improve it. Characteristics of the organizational culture were obtained according to the suggested basic parameters from the standpoint of the management and employees of enterprises.

Keywords: organizational culture, expert evaluation, ranking, modeling, analysis, competitiveness

The method of design and increments in solving linear programming problems

2022. T.10. № 3. id 1223

Ganicheva A.V. Ganichev A.V.

DOI: 10.26102/2310-6018/2022.38.3.022

Currently, the issue of choosing the optimal solution is one of the most important and urgent in industry, economy, agriculture, and the military sector. Methods and approaches of linear programming theory are used to solve many applied optimization tasks. The simplex method, which is the principal method of linear programming, is characterized by a large amount of computational actions and procedures. Owing to this, modifications of the main method with higher algorithmic efficiency are employed to address this problem. In this article, a new method for solving linear programming problems has been developed. The algorithmic complexity, which is less than that of the simplex method, is provided by considering a class of problems with completely limited areas of acceptable solutions. The new method is justified by the results announced in the proven statements. The implementation of the method is described by two algorithms: 1) search for a quasi-optimal solution by analyzing the coordinates of projections on hyper planes (design algorithm); 2) search for an optimal solution by setting increments to constraints (increment algorithm). To explain the functioning of the algorithms, specific numerical examples are analyzed. Algorithmic complexity estimates of the developed method are carried out by counting the number of arithmetic operations undertaken. Formula expressions for estimating the complexity of calculations are obtained.

Keywords: algorithm, variable, hyperplane, projection, inequality, iteration, number of operations, computational complexity

Classification of random signals based on their doubly connected Markov models

2022. T.10. № 3. id 1222

Kalinin M.Y. Choporov O.N. Bonch-Bruevich A.M.

DOI: 10.26102/2310-6018/2022.38.3.017

The article considers the problem of identifying the pre-selected class of an observed signal. This appears to be a relevant issue in the theory of pattern recognition, clustering, statistical decisions, technical diagnostics, and a number of other areas of science and technology. As a signal model, its doubly connected Markov model (complex Markov chain) is used based on three-dimensional probability densities of simulated random processes. The technique for forming class models according to known probabilistic characteristics or according to a classified training sample of samples is regarded. As a part of the Bayesian approach, the posterior probabilities that determine the affiliation of the observed sample of signal samples with each class are defined. An optimal signal classification algorithm is proposed, a decision-making algorithm is developed, decisive statistics are formed that depend on the observed sample of samples and matrices of transition probabilities of the analyzed classes, providing means for decision-making with a given reliability and based on the Wald procedure; their properties are also examined. Statistical simulation of the classification algorithm has been carried out, which confirms its effectiveness. The research results can be used in various systems and devices for detecting objects according to the random signals generated by them, for example, in technical diagnostics equipment.

Keywords: signal, classification, markov model, wald procedure, decision statistics

Ships route planning in heavy-traffic marine area based on historical data

2022. T.10. № 3. id 1221

Grinyak V.M. Devyatisilnyi A.S.

DOI: 10.26102/2310-6018/2022.38.3.014

The paper considers the problem of planning a route for sea vessel shifting. Under the conditions of heavy traffic, navigators should follow the traffic scheme accepted in this defined water area. Such a pattern may not be officially established while representing collective experience in navigation. In this case, route planning based on the data on the movement of other ships that had been in this water area before (the same idea underlies the methods of "big data" tasks) appears to be productive. In the papers published earlier, such route planning employed a cluster analysis of retrospective data on the movement of ships, which involved dividing the water area into sections and isolating their characteristic values of speeds and courses. The problem with this approach was the choice of partitioning parameters, which had to be set for each specific water area separately. This paper proposes another approach when the graph of potential routes includes a selection of the trajectories of individual ships that had been previously implemented in the selected water area. The article regards a method for constructing such a graph of possible routes, estimates the number of its vertices and edges, and gives recommendations on the choice of a method for finding the shortest path on this graph. A possible method premised on the notion of combining straight and maneuverable sections of vessel traffic that can be applied to interpolate the missing data required to build a graph is discussed. Examples of route planning in a number of real water areas are given: Vladivostok, Tokyo Bay, the Tsugaru Strait.

Keywords: maritime safety, route planning, big Data, automatic identification system, graph algorithms, shortest path

Mathematical model of salt ion stationary transport in the cross section of the channel at equilibrium

2022. T.10. № 3. id 1219

Chubyr N.O. Kovalenko A.V. Urtenov M.K. Gudza I.V.

DOI: 10.26102/2310-6018/2022.38.3.009

The equilibrium at the interphase boundaries largely determines the transfer processes and therefore studying it is an important task. The paper proposes a mathematical model of the problem of salt ion stationary transfer at the onset of equilibrium, namely at zero current, in the cross section of the desalination channel formed by anion exchange and cation exchange membrane in the form of a boundary value problem for systems of Nernst-Planck and Poisson equations in the potentiostatic mode. A numerical and asymptotic solution of this boundary value problem is obtained. The numerical and asymptotic solutions are compared, and their coincidences were shown with good accuracy. The acquired asymptotic solution allows for an exhaustive analysis of the equilibrium state depending on the initial concentration, potential jump, and properties of ion-exchange membranes and helps to establish the basic transfer patterns. It is shown that the stationary state of salt ion transfer process through the channel section coincides with the equilibrium state. The location and dimensions of the spatial charge and electroneutrality regions are established. The dependence of the electric field strength and concentration on the potential jump and the boundary values for cation and anion concentrations is obtained. The results of the research can be used to determine the optimal operating modes of electrodialysis water purification devices.

Keywords: small parameter, asymptotic solution, cross section of desalination channel, electromembrane systems, numerical solution, singularly perturbed problems

The use of an artificial neural network in thermal diagnostics of the printed node of the on-board take-off control device of an aircraft

2022. T.10. № 3. id 1218

Uvaysov S.U. Chernoverskaya V.V. Dang N.V. Tuan N.V.

DOI: 10.26102/2310-6018/2022.38.3.012

Technical diagnostics and monitoring of an electronic device are integral parts of its life cycle since they help to assess not only the technical condition of components and modules in real time, but also make it possible to identify hidden defects that have arisen during the production or operation of the device, and make a forecast about the residual life of the product. It can be said that in the process of technical diagnostics, the reliability indicators of the device under study and the compliance degree of the embedded and implemented functionality are evaluated, which is inextricably linked with the qualitative characteristics of the product. Modern radio-electronic devices characterized by high circuit, structural and technological complexity require additional study of the existing diagnostic methods and the search for new approaches to increasing the resolution, reliability, and effectiveness of diagnostic procedures. In this area, achievements from the field of artificial intelligence, machine learning, and neural networks along with traditional, proven methods have been actively used recently. In addition, the use of modeling and computational experiment in design made it possible to combine design and diagnostic procedures, conduct diverse studies of the virtual twin of the device and make the necessary changes in a timely manner, thereby preventing the manifestation of negative effects in the finished product at the early stages of development even before the production of a prototype. The article presents the results of a study aimed at creating a thermal model of the designed node and developing an artificial neural network for recognizing structural defects of the device by its thermal field. In this research, specialized computer-aided design systems were actively employed, including engineering analysis and calculation tools, as well as the high-level Python programming language. The findings have a practical importance and can be utilized by developers of radio-electronic devices in order to achieve high reliability and operational characteristics of the product at all stages of its life cycle.

Keywords: radio-electronic device, on-board device, control of aircraft take-off dynamics, thermal mode, thermal modeling, artificial neural network, fault database, computer-aided design system, technical diagnostics

Modified genetic algorithm for project scheduling

2022. T.10. № 3. id 1214

Korotkov V.V.

DOI: 10.26102/2310-6018/2022.38.3.007

The paper describes a modified genetic algorithm for solving resource-constrained project scheduling problem. The relevance of the study is due to the widespread prevalence of project organization of activities and the extremely high computational complexity of the problem under consideration. Further improvement of existing heuristic algorithms is needed to enable efficient planning of large projects. The available genetic algorithms are based on activity order encoding methods and implementations of genetic operators, which does not fully take into account the specifics of the problem. Therefore, the paper proposes an alternative encoding method and the corresponding crossover operator, which, unlike classical approaches, highlights relative rather than absolute positions of activities as inherited features. The study regards the main properties of such encoding which can be represented as square Boolean matrices. A mapping operator that helps to reduce Boolean matrices to a canonical row form is also introduced. The resulting genetic algorithm and classical implementations were compared using a test set of tasks. The suggested approach has shown potential efficiency, especially with large projects. The findings can be of practical importance in the development of decision support systems for project management.

Keywords: genetic algorithm, crossover operator, project planning, combinatorial optimization, scheduling theory

The impact of coronavirus infection on the socio-economic indicators of the region

2022. T.10. № 3. id 1213

Pecherina A.V.

DOI: 10.26102/2310-6018/2022.38.3.028

The new coronavirus infection (COVID-19) which emerged in Wuhan, China, in early December 2019 quickly spread to almost every country in the world and shocked the global economy. This article highlights the most important problems that are caused by the coronavirus pandemic. The author discusses the impact of the new coronavirus infection Covid-19 on some socio-economic indicators of a particular region of the Russian Federation as well as the Russian Federation as a whole. In order to do that, an analytical procedure was developed using Knime Analytics Platform (the free and open source data analysis platform), which, in turn, greatly simplified data processing and visualization of results. The platform makes it possible to develop reproducible and scalable workflows by integrating a wide range of analysis tools. The analysis was based on the data extracted from the website of the Center for Spatiotemporal Innovation at Harvard University (NSF Spatiotemporal Innovation Center) and the statistical data extracted from the website of the Federal State Statistics Service. We visualized the data and drew conclusions about COVID-2019 incidence rate and the cost of a constant set of consumer products and services for the purposes of inter-regional comparisons of purchasing power.

Keywords: data analysis, data mining, covid-19, coronavirus infection, socio-economic indicators

Detection of information security threats using deep neural networks in computer networks in real time

2022. T.10. № 3. id 1212

Trunov E.E. Klyuev S.G.

DOI: 10.26102/2310-6018/2022.38.3.011

Currently, the issue of detecting information security threats in computer networks is becoming a problem when it comes to preventing such threats in real time. The number of subscribers of almost any computer network is growing and so does the number of threats that can create a potential danger to the functioning of the network. In this regard, modern mechanisms that will help to respond to emerging information security threats in a timely manner are required. In this paper, the analysis of possible mechanisms of protection against security threats in computer networks is carried out and a methodology for implementing such protection using neural networks is proposed. In addition, a control example is implemented with a trained deep neural network which is able to detect information security threats with high accuracy and minimal delays. The materials of the article are of practical value when incorporating such a neural network into an intrusion detection system. By means of the method proposed in the article, it is possible to achieve a near-real-time response to information security threats and, as a result, prevent possible information security accidents.

Keywords: computer network, neural network, security threat, deep learning, protection mechanism

Ensemble methods for detecting outliers in the preparation of a training data set

2022. T.10. № 3. id 1210

Dorofeev V.S. Volosatova T.M.

DOI: 10.26102/2310-6018/2022.38.3.013

Most machine learning methods are most effective when working with data that satisfies a nor-mal distribution. On the other hand, the training set often contains “outliers” of various nature, which can significantly reduce the accuracy of machine learning methods. Thus, in any machine learning task, there is a problem of detecting outliers. The article provides a classification of the main types of emissions. Various methods for detecting one-dimensional outliers are considered: the method using the Grubbs criterion; Z-score method; robust Z-score (RZ-score) method; in-terquartile range (IQR) method; Winsorization method. The methods for detecting one-dimensional outliers are compared. For the automated detection of outliers, an ensemble method has been proposed that combines various methods for detecting one-dimensional outliers. The ensemble method helps to configure an automated outlier detection procedure according to the rule of the required severity. The suggested method is applied to analyze and detect outliers in data on sales of goods during the promotion in a large retail network. The applicability of using outlier detection method ensemble to stratification of the training sample is shown. At the same time, the absolute and relative forecasting error of the final model decreased by 5% compared to the initial one.

Keywords: outliers, machine learning, training sample, ensemble method, z-score, interquartile range method

Vessels route planning under ice conditions

2022. T.10. № 3. id 1209

Grinyak V. Akmaykin D. Ivanenko Y.

DOI: 10.26102/2310-6018/2022.38.3.006

This paper is devoted to the problem of navigation safety in ice-covered sea areas. The route planning is examined as a means of lessening the impact of ice upon a vessel as it follows its course, taking into consideration the type of ice and the ice class of the vessel. Special information services based on satellite monitoring as well as reports from vessels and polar stations, presenting information on ice cover in tables and diagrams, can be used as the source of data on the ice situation along the route. The current paper proposes a pattern concept of route planning and notes the complexity of its implementation. Simplifying the problem by finding the shortest way of the route in the weighted graph is suggested, which is a conventional strategy in ship navigation. Possible approaches to developing a set of graph nodes and edges as well as weighing the graph edges are discussed. Some recommendations for reducing computational complexity of tasks are given. The paper is accompanied with calculations of vessel routes using the data on ice situation in the sea of Okhotsk. The given examples show that the ship's track is formed in such a manner that the traffic in the sea areas covered by ice is decreased. Following on from the results of calculating routes under various ice conditions, a conclusion is made about the possibility of solving the problem in this way.

Keywords: navigation safety, route planning, ice conditions, vessel ice class, graph algorithms, shortest path

Synthesis of cognitive-constructive process management in human-technical-natural systems

2022. T.10. № 3. id 1208

Beltiukov A.P. Maslov S.G.

DOI: 10.26102/2310-6018/2022.38.3.005

The paper deals with the problems of synthesizing emotionally oriented management of cognitive-constructive processes in human-technical-natural systems with respect to the influence of heterogeneous illusions and misconceptions. The general problem statement is examined, logical models that reflect the task are proposed. The basic foundations of the synthesis of various factors for solving problems are identified. Emotions, illusions, and misconceptions are considered among such factors. This is done to overcome emerging contradictions and address the problems with due regard for the positive and negative aspects of the listed factors. Another goal of this is the formation of an adequate environment and means of cognitive and constructive activity in terms of complex system specifics, including humans, technology, and natural objects. The foundations for a new systematization of factors and adequate means of cognitive-constructive activity are being created. These basics take into account the influence of emotions, illusions and misconceptions. Individual and collective characteristics of intellectual and information resources are analyzed along with the features of objective and subjective aspects of control synthesis. New abstract and specific models are being designed that improve cognitive and constructive activity, the development and use of knowledge for the problems being solved.

Keywords: emotionally oriented management, synthesis, human-technical-natural system, cognitive-constructive process, emotions, illusions, misconceptions

Fractal calculus application for analysis of the results of an associative-verbal experiment

2022. T.10. № 3. id 1206

Barinov V.R. Philippovich Y.N.

DOI: 10.26102/2310-6018/2022.38.3.018

The paper supports the hypothesis on the use of fractal calculus to process the results of an associative-verbal experiment. The results of associative-verbal experiments are unstructured data with a large volume. The relevance of the study is due to the fact that the existing methods of processing the results of associative-verbal experiments cannot be unified because the data obtained have significant differences in formats. In turn, this limits the possibilities of creating and applying typical data processing algorithms for various associative-verbal experiments. One of the main problems of computational linguistics is the heterogeneity of hierarchical structures describing human consciousness and speech, as well as their constant modification. One of the possible solutions to this problem is to employ the method of fractal calculus that can help develop a model of linguistic consciousness having not only a complex structure of its own but is in constant interaction with other structures of the real world. Leading Russian scientists have substantiated means for addressing similar issues with unstructured data in other academic fields, which is analyzed in this article. Based on the presented research of scientists, a comparative analysis of existing fractal models for solving problems in various fields of activity is given, and a hypothesis about the possibility of using this method to model linguistic consciousness is suggested.

Keywords: fractal, calculus, associative-verbal experiment, artificial intelligence, big data, language consciousness, information search

Methods of user identification in the information and telecommunication environment based on the analysis of account attributes

2022. T.10. № 3. id 1203

Romanov A.G.

DOI: 10.26102/2310-6018/2022.38.3.002

The relevance of the study is due to the problem of the growing number of unidentified persons who have committed crimes on the Internet and beyond. In this regard, the aim of the article is to demonstrate the means for personal identification by identifying users in the virtual space in order to convict them of criminal offence. The improvement of information technologies and the development of services in the information and telecommunications space provide an opportunity to analyze numerous data, including those left by users about themselves in social networks. Thus, the leading method to investigate the problem is the techniques to determine the similarity of alphanumeric objects created by users in the attributes of social network profiles. This article presents a possible algorithm of actions to deanonymize the identity of a criminal. The development and application of methods for identifying users in the virtual space will allow us to comprehensively consider the existing problem and accomplish one of the main tasks assigned to the internal affairs bodies and related to crime solving and charging perpetrators with a criminal offence. The materials of the article may be of practical value to the internal affairs bodies in the terms of enhancing the efficiency and effectiveness of law enforcement activities.

Keywords: user identification, internet, data analysis, social networks, crimes

On the approach to forecasting indicators of socio-economic development of the region based on indirect indicators

2022. T.10. № 3. id 1202

Rusanov M.A. Abbazov V.R. Baluev V.A. Burlutsky V.V. Melnikov A.V.

DOI: 10.26102/2310-6018/2022.38.3.004

Economic and social development requires constant modernization of the regional management system based on the system of key socio-economic indicators of the region's development and methods of their analysis and forecasting. The article proposes a comprehensive approach to forecasting based on the application of classical forecasting methods for existing time series of statistical indicators and by identifying and analyzing indirect semantically close indicators to a new indicator in the absence of the necessary time series for forecasting. The article provides a general methodology for obtaining a forecast and describes in detail the method for constructing a forecast estimate of the change dynamics in the estimated indicator as well as a description of the AutoML library with open source FEDOT, which was used to build forecasts. The issue of constructing and optimizing a combined forecast with the aid of automatic machine learning tools is considered. At the end of the article, the result of an experiment on predicting the indicators “Population of the subject of the Russian Federation” and “Life expectancy at birth” according to the proposed approaches and a comparison of the findings is presented. It can be concluded that the suggested approach to making a predictive assessment of the change dynamics in the estimated indicator by identifying indirect indicators can be applied to socio-economic indicators of the development of the region.

Keywords: socio-economic indicators, forecasting, incompleteness, autoML, indicator of senior official activity effectiveness

Development of a method for determining the dominant type of human breathing pattern based on computer vision technologies, motion capture systems and machine learning

2022. T.10. № 4. id 1200

Zubkov A.V. Donskaya A.R. Busheneva S.N. Orlova Y.A. Rybchits G.M.

DOI: 10.26102/2310-6018/2022.39.4.016

The study raises the problem of the absence of methods for determining the dominant type of breathing pattern that can be used in the implementation of software products that contribute to the support of patients with respiratory insufficiency and their rehabilitation at the stage of inpatient and outpatient treatment. Existing methods are either too labor-intensive to implement due to the excessive amount of markers utilized by motion capture systems or economically unprofitable due to the cost of the equipment itself or developed only for research purposes and are not applicable in clinical practice. In this regard, this article is aimed at developing a method for determining the type of breathing, which could later be employed for automated rehabilitation of patients with respiratory insufficiency. As part of the study, computer vision and machine learning methods were applied as well as methods based on motion capture technologies. The article presents methods for determining the position of markers in space and analyzing the type of human breathing (thoracic, abdominal, mixed) in real time based on the data obtained by means of motion capture system markers. The materials of the article are of practical value in the field of medical rehabilitation of patients with respiratory insufficiency; they make it possible to optimize labor processes within the field of medical rehabilitation, i.e. reducing labor and time costs of rehabilitologists.

Keywords: computer vision, neural networks, motion capture systems, patient rehabilitation systems, detection of breathing patterns

Modified Elman neural network with dynamic learning rate for tracking and motion prediction of a nonholonomic three-wheeled mobile robot

2022. T.10. № 3. id 1199

Berezina V.A. Mezentseva O.S. Mezentsev D.V.

DOI: 10.26102/2310-6018/2022.38.3.003

This article proposes to track and predict the trajectory of a non-holonomic three-wheeled mobile robot using a modified Elman neural network. An algorithm for calculating the learning rate of a neural network is suggested, which improves the efficiency and speed of learning and also reduces the number of iterations required for learning. The modified Elman algorithm with dynamic learning rate (MENN) is compared with the classical Elman neural network (ENN) and the PSO algorithm (PSO-ENN). Training of a neural network is evaluated according to two criteria: the number of iterations required for training and the average training time. In addition, the deviation from the given trajectory of movement is checked: movement along a straight line, in a square and in a circle for each algorithm. The simulation results showed that the modified Elman neural network with a dynamic learning rate is more efficient (by 32.4% on average) and accomplishes the learning objective faster (by 66.4% on average) and has the least deviation from the given motion trajectory. The relative measurement error ranges from 7.8% to 20.2% at 95% reliability and five tests for each group of measurements.

Keywords: recurrent neural network, elman neural network, learning rate, nonholonomic three-wheeled robot, motion trajectory prediction

Modeling of artificial intelligence system for early detection of emergency situations at vital facilities

2022. T.10. № 3. id 1197

Borovskoy I.G. Shelmina E.A. Afanasyeva I.G. Matolygin A.A.

DOI: 10.26102/2310-6018/2022.38.3.001

The article presents the results of modeling an artificial intelligence system for early detection of undesirable situations of various types at objects of particular national economic importance. Pipeline transport or any other production system, in which continuous monitoring of operability parameters of critical components and mechanisms is carried out, can be specified as such object. This model can be applied by various oil and gas production companies. The results of modeling and subsequent development of the information system will provide the basis for industrial implementation of highly effective systems of accident detection and prevention in reliance on neural network analysis of continuously received streaming data. As a part of this research, the possibility of using modern neural network architectures for the problem under consideration is examined, namely, convolutional neural networks – TCN, direct propagation neural networks – MLP, recurrent neural networks – LSTM. It was proposed to abandon the activation function for LSTM which helps to provide the neural network with "long-term memory" of stored values, which is crucial to this problem. In addition, a cross-comparison of the error reduction rate during network training was performed to detect an architecture capable of "self-learning". All models were tested with the aid of the training data from the "Vostochny kupol" wells. Acceptable coincidence of test and extrapolation data was obtained for all models.

Keywords: artificial intelligence, time series, artificial neural network, emergency, neural network architecture, convolutional neural networks, direct propagation neural networks, recurrent neural networks

Development of a steganalysis system for digital images based on a neural network classifier

2022. T.10. № 2. id 1196

Minaychev A.A. Mezentsev A.O. Yandashevskaya E.A.

DOI: 10.26102/2310-6018/2022.37.2.020

The article discusses an approach to the implementation of a system for steganographic analysis of digital images based on a neural network classifier. It is used as a part of an integrated system for monitoring information security events of corporate infocommunication systems. As a basic structure for the neural network classifier, it is proposed to use a modified version of the convolutional neural network. Its preprocessing module implements the histogram method for analyzing the color and brightness characteristics of digital images. To automate the learning process of the neural network classifier, it is suggested to introduce a module for mass generation of stegocontainers with predefined values for the type and size of a digital image as well as for the size of the payload into the structure of the system being developed. Based on the developed structure of the steganalysis system for digital images, a factorial experiment was planned and conducted to evaluate the quality of the described neural network classifier in comparison with the known solutions of binary statistical classifiers. The choice of the area under the error curve (AUC ROC) as a metric for assessing the quality of classification is the main feature of the experiment. The results show that it is possible to use neural network classifiers to solve steganalysis problems, including their implementation in advanced information security tools.

Keywords: digital steganography, digital images, convolutional neural network, binary classification, steganographic container, classification accuracy

Support decision-making for analyzing the effectiveness of a website using Web Usage Mining methods

2022. T.10. № 2. id 1191

Zelenina A.N. Kokorina A.I. Petrosov D.A.

DOI: 10.26102/2310-6018/2022.37.2.019

In the modern world, one of the most effective methods to maintain the functioning of an organization or business with a view to facilitating development is to design a website and then to employ it to communicate with users and customers. The website helps to systematize all information about the organization, provides a means of e-commerce and gives the opportunity for representatives of the organization and users to communicate with each other to exchange ideas or feedback on products or services. Thus, effectiveness analysis of the website and appropriate decision-making, regarding its optimization and changes to the design, which will allow the company subsequently to achieve its goals, becomes more relevant. In this article, a decision support system was implemented to analyze the effectiveness of a website using Web Usage Mining methods. Statistical methods, which enable performance improvement of the website based on the information received, were chosen as well as data mining methods, in particular, clustering and association rules that are utilized to personalize content and, in the case of selling websites, purchasing offers, which will significantly increase the loyalty of users and customers.

Keywords: decision support system, web Usage Mining, website, log file, machine learning, clusterization, association rules