The authors present a method for detecting anomalies in the visual content of critical information infrastructure based on comparing hash strings obtained from visual data to detect potential deviations or duplicate content. The subject of the study is the detecting of anomalies in critical information infrastructure (CII) images using hashing technology. The relevance of the study is due to the growing threat of copyright infringement and the distribution of illegal content. Critical information structures, including public administration, science, economics and energy systems, directly depend on the protection of their information. Therefore, identifying anomalies in images and their prompt response play a key role in maintaining the integrity and confidentiality of data. The goal is to develop an algorithm for identifying duplicate content and create an effective tool for monitoring images. The work uses the integration of computer vision methods and machine learning algorithms. The development includes the use of hash strings for precise comparison of images. The scientific novelty of this study lies in the development and implementation of a new approach to detecting anomalies in images of critical information infrastructure using hashing technology. The use of the technology provides unique identifiers for visual data and allows for efficient comparison and analysis of images. This approach significantly increases the speed of data processing and the accuracy of detecting duplicate content and anomalies in images. Hash-based image classification provides a higher degree of sensitivity to anomalies and allows filtering out false positives, which is critical for organizations with a high level of information security. The results show the high efficiency of the proposed method; a significant degree of accuracy in detecting anomalies has been achieved, which is confirmed by experiments on real data. The presented algorithms have demonstrated improvement over existing solutions.

Critical information infrastructure (CII), including the financial sector, plays a key role in ensuring the sustainable functioning of economic systems and the financial stability of States. However, the growing digitalization of the financial industry and the introduction of innovative technologies are opening up new attack vectors for attackers. Modern cyber attacks are becoming more sophisticated, and traditional defenses are proving ineffective against new, previously unknown threats. There is an urgent need for more flexible and intelligent cybersecurity systems. Thus, the subject of the study is modern intelligent methods and technologies for protecting critical information infrastructure (CII) from cyber attacks. The object of the research is methods and means of ensuring the protection of critical information infrastructure using artificial intelligence and machine learning technologies. The methodological basis of this study is a comprehensive analysis of the scientific literature on the use of intelligent methods and technologies to protect critical information infrastructure. During the review and critical analysis of relevant scientific publications, key problems and unresolved tasks requiring further scientific research and practical developments in this subject area were identified. This methodological approach allowed us to form a holistic view of the current state and prospects for the development of intelligent cybersecurity tools for critical financial systems, as well as to identify priority areas for further research. The main directions of scientific novelty of this research are: 1. A detailed review of promising approaches based on artificial intelligence and machine learning technologies to ensure effective protection of CII organizations from modern complex cyber attacks. 2. Identification and analysis of a number of key scientific and technical problems that need to be solved to increase reliability, interpretability and trust in intelligent cybersecurity systems, including issues of robustness to attacks, active online learning, federated and differential private data processing. 3. Identification of promising areas for further research and development in the field of application of specialized methods of secure and trusted AI to protect critical financial infrastructure. Thus, this research makes a significant contribution to the development of scientific and methodological apparatus and practical solutions for the use of intelligent methods to ensure cybersecurity.

At the moment, neural networks are able to create what was previously considered inaccessible: photorealistic faces of people; full-fledged paintings based on rough sketches; any images with a short text description; poems and prose on the first lines or a given topic. All of this has been made possible by rapid advances in areas such as natural language processing and machine vision. Neural networks are capable of generating content based on the data they have memorized during an extensive learning process. Problems for logic, mathematics and logical reasoning are an example of flexible intelligence, and it requires completely different approaches to learning. The study presented in the article proposes the development of a methodology for designing and training neural networks aimed at creating a functioning code. The basis of the study is the possibility of using artificial intelligence, in particular neural networks, to generate code by a machine, that is, AI4Code tasks. The study examined the provisions in favor of the use of reinforcement learning in comparison with language models, as well as the architecture of the environment necessary for such learning. The main research is the focus on the use of Nvidia graphics accelerators and the use of central processes of various architectures. The article discusses the features of creating a learning environment, the advantages and disadvantages of the CUDA platform, and analyzes the potential effectiveness of each of the approaches.

Машинное обучение позволяет такому банковскому бизнесу, как кредитование, нести минимальные потери.Так, правильная классификация заемщика при его обращении с запросом на получение кредита сводит вероятность предоставления банком денежных средств под проценты сомнительному клиенту к минимуму, а безопасному -к максимуму.Такой исход, в свою очередь, гарантирует как преувеличение капитала за счет выдачи займов тем, кто их сможем погасить, так и его сохранение благодаря возможности применить «фильтр» и избежать кредитования ненадежных клиентов.Подобное стало реализуемо благодаря решению задачи, которое получило название «кредитый скоринг».В данной статье будут рассмотрены три технологии, реализованные авторами для достоверной классификации заемщиков: Artificial neural network, XGBoost classifier и Random forest classifier.Применимость всех методов была проверена экспериментально на реальных данных

The research is aimed at solving the problem of the execution of government contracts, the importance of using unstructured information and possible methods of analysis to improve the control and management of this process. The execution of government contracts has a direct impact on the security of the country, its interests, economy and political stability. Proper execution of these contracts contributes to the protection of national interests and ensures the security of the country in every sense. The object of research is algorithms used to extract information from texts. These algorithms include machine learning technologies and natural language processing. They are able to automatically find and structure various entities and data from government contracts. The scientific novelty of this study is the accounting of unstructured information in the analysis of the execution of government contracts. The authors drew attention to the problem-oriented texts in the contract documentation and suggested analyzing them with numerical indicators to assess the current state of the contract. Thus, a contribution was made to the development of methods for analyzing government contracts by taking into account unstructured information. The proposed methods for analyzing problem-oriented texts using machine learning. This approach can significantly improve the evaluation and management of the execution of government contracts. The results of the interpretation of problem-oriented texts can be used to optimize the risk assessment model for the execution of a government contract, as well as to increase its accuracy and efficiency.

Рассматривается вопрос внедрения STEM-образования в высшую школу в России. Автором рассмотрены исторические предпосылки поиска новых образовательных технологий и их ис-пользование в зарубежных странах. Определено влияние уровня образования на экономическую и производственную сферы деятельности государства.

Развитие технологий влияет на трансформацию всех сфер деятельности, отраслей экономики — тренд на минимизацию участия человека в процессах, их автоматизации, оптимизацию работы и освобождению времени. Не исключение и сфера образования, где с переносом процессов в онлайн-формат возросла необходимость в инструментах и сервисах и донесения знаний через интернет. При этом стоит отметить возрастающую популярность именно мобильных устройств. В данной статье представлен обзор возможностей чат-ботов на базе социальной сети «ВКонтакте» и их применение в сфере образования. Были проанализированы статистические данные «ВКонтакте» и Telegram, а также распространение мобильных устройств для обоснования актуальности использования данной технологии. Рассмотрены существующие чат-боты, применяемые в образовательном процессе. Проанализированы задачи, которые доступны для автоматизации через чат-ботов, направленные на повышение эффективности выполнения данных задач. Помимо этого, рассмотрена технологическая специфика создания чат-ботов на базе «ВКонтакте». The development of technologies affects the transformation of all spheres of activity, sectors of the economy - a trend to minimize human participation in processes, automate them, optimize work and free up time. The field of education is no exception, where with the transfer of processes to the online format, the need for tools and services and the delivery of knowledge via the Internet has increased. At the same time, it is worth noting the growing popularity of mobile devices. This article provides an overview of the capabilities of chatbots based on the social network VKontakte and their application in the field of education. The statistical data of VKontakte and Telegram were analyzed, as well as the distribution of mobile devices to justify the relevance of using this technology. Existing chatbots used in the educational process are considered. Analyzed the tasks that are available for automation through chatbots, aimed at improving the efficiency of their implementation. In addition, the technological specifics of creating chatbots based on VKontakte are considered.

Развитие технологий влияет на трансформацию всех сфер деятельности, отраслей экономики — тренд на минимизацию участия человека в процессах, их автоматизации, оптимизацию работы и освобождению времени. Не исключение и сфера образования, где с переносом процессов в онлайн-формат возросла необходимость в инструментах и сервисах и донесения знаний через интернет. При этом стоит отметить возрастающую популярность именно мобильных устройств. В данной статье представлен обзор возможностей чат-ботов на базе социальной сети «ВКонтакте» и их применение в сфере образования. Были проанализированы статистические данные «ВКонтакте» и Telegram, а также распространение мобильных устройств для обоснования актуальности использования данной технологии. Рассмотрены существующие чат-боты, применяемые в образовательном процессе. Проанализированы задачи, которые доступны для автоматизации через чат-ботов, направленные на повышение эффективности выполнения данных задач. Помимо этого, рассмотрена технологическая специфика создания чат-ботов на базе «ВКонтакте».

The subject of the research is assessing the risks of performing government contracts. The object of the study is the process of analysis and evaluation of the implementation of government contracts. The study is aimed at developing a methodology that determines the importance and significance of signs that influence the risk of non-fulfillment of government contracts. Research methods were used: data analysis to detect connections and dependencies between various characteristics and the risk of non-fulfillment of government contracts; statistical analysis to obtain an assessment of the impact of each characteristic on the risk of non-fulfillment of contracts and ranking them in order of importance; machine learning to predict the risk of non-fulfillment of government contracts; expert assessments to take into account contextual factors and features, their impact on the importance of features. The main conclusions of the study are the presented methods for assessing the importance of features when analyzing the implementation of government contracts, by using data from various sources, including the register of public procurement of the unified information system (UIS), the register of unscrupulous suppliers (RNP) of the EIS and the SPARK information system. The authors managed to achieve high prediction accuracy (more than 97%) and analyze the most important and significant features. The scientific novelty lies in the fact that the results obtained make it possible to identify and analyze factors from three information systems that influence the risks of non-fulfillment of government contracts. Thus, this study is valuable and important in its field, which contributes to the development of more effective risk management methods and increased efficiency in the implementation of government contracts. The results obtained allow us to identify the factors that have the greatest impact on the risks of non-fulfillment of contracts, which makes the study valuable and important in this area.

The methodology for evaluating the execution of government contracts in the energy sector by means of machine learning is presented. The signs describing performers and customers in the public procurement system were identified, the risks of fulfilling contracts on the part of customers and performers were identified, the main categories for compiling a dataset were identified and a dataset was assembled. Data problems are described, and ways to fix these problems are described. The problem of classifying the execution of government contracts is solved, a software package for intelligent forecasting of the execution of government contracts is described

The power industry plays a key role in ensuring the energy security of the state. Sustainable and reliable functioning of the energy system requires the fulfillment of contracts with strictly observed deadlines and quality of work. This article describes an algorithm for selecting methods for interpreting machine learning models, analyzes gradient boosting-based machine learning methods recommended for solving prediction tasks in the field of power engineering, and presents methods for interpreting the results. The authors have achieved good results in training models and determined objective assessments of the contribution of each feature to solving the prediction tasks of contract fulfillment. This research is significant in the context of ensuring the efficiency and transparency of public procurement and can be beneficial for specialists and government bodies responsible for monitoring contract fulfillment in the field of power engineering.

На текущий момент нейросети способны создавать то, что раньше считалось недосягаемым: фотореалистичные лица людей; полноценные картины по грубым наброскам; любые изображения по краткому текстовому описанию; стихи и прозу по первым строчкам или заданной теме. Все это стало возможно благодаря стремительному прогрессу в таких областях как обработка естественного языка и машинное зрение. Нейросети способны генерировать контент на основе тех данных, которые они запомнили во время обширного процесса обучения. Задачи на логику, математику и логические рассуждения является примером гибкого интеллекта, и он требует совершенно других подходов к обучению. Представленное в статье исследование предлагает выработку методологии проектирования и обучения нейросетей направленных на создание функционирующего кода. Основой исследования является возможность применения искусственного интеллекта, в частности нейронных сетей, на генерацию кода машиной, то есть задач AI4Code. В исследовании рассмотрены положения в пользу применения обучение с подкреплением в сравнении с языковыми моделями, а также архитектура необходимой для такого обучения среды. Основной исследования является направленность на использование графических ускорителей Nvidia и использование центральных процессов различных архитектур. В статье рассмотрены особенности создания среды обучения, достоинства и недостатки платформы CUDA, проведен анализ потенциальной эффективности каждого из подходов. At the moment, neural networks are able to create what was previously considered inaccessible: photorealistic faces of people; full-fledged paintings based on rough sketches; any images with a short text description; poems and prose on the first lines or a given topic. All of this has been made possible by rapid advances in areas such as natural language processing and machine vision. Neural networks are capable of generating content based on the data they have memorized during an extensive learning process. Problems for logic, mathematics and logical reasoning are an example of flexible intelligence, and it requires completely different approaches to learning. The study presented in the article proposes the development of a methodology for designing and training neural networks aimed at creating a functioning code. The basis of the study is the possibility of using artificial intelligence, in particular neural networks, to generate code by a machine, that is, AI4Code tasks. The study examined the provisions in favor of the use of reinforcement learning in comparison with language models, as well as the architecture of the environment necessary for such learning. The main research is the focus on the use of Nvidia graphics accelerators and the use of central processes of various architectures. The article discusses the features of creating a learning environment, the advantages and disadvantages of the CUDA platform, and analyzes the potential effectiveness of each of the approaches.

The subject of the research is the development of a software package for intelligent forecasting of the execution of government contracts using machine learning methods and analysis of unstructured information. The object of the study is the process of control and decision-making in the field of public procurement, including the selection of contractors, the execution of contracts and the assessment of the timing and cost of their implementation. Special attention in the study is paid to the development and application of interpreted machine learning methods to solve the problems of assessing the risks of choosing an unscrupulous contractor, the risks of non-fulfillment of the contract on time and forecasting the likely timing and cost of contract implementation. The authors consider in detail such aspects as a unique set of data that was collected from various information systems. They have also developed automated data collection and update systems that can be installed on customers' servers. The methods of machine learning, analysis of unstructured information and interpreted methods were used in the work. Interpreted machine learning models were built to assess the risk of choosing an unscrupulous contractor, assess the risk of non-fulfillment of the contract on time, as well as assess the likely timing and cost of contract implementation. A unique set of data was collected in the work, including more than 83 thousand data on more than 190 features from various systems, such as the Unified Information System (UIS) Public Procurement Register, the Register of Unscrupulous Suppliers (RNP) EIS and SPARK Information System. Automated data collection and updating systems have been developed that can be deployed on customer servers. In the course of the study, software packages were developed for intelligent forecasting of the execution of government contracts, which provide an opportunity to conduct a more accurate risk analysis using unstructured information analysis methods, machine learning models and interpreted methods. This makes it possible to increase the effectiveness of monitoring the implementation of government contracts and reduce the likelihood of corruption and violations. The study demonstrates the importance and applicability of machine learning methods and models in the field of public contracts and provides new opportunities for improving control and decision-making processes in the field of public procurement.

The quality of knowledge is the most important task of the learning system at any level and stage of education. The COVID-19 pandemic has made its own adjustments to the process of organizing education, requiring the transition from traditional to distance learning as soon as possible. In the new conditions, the use of adaptive knowledge control has become relevant, taking into account the individual level of knowledge of the trainees. The study is devoted to the analysis of the features of adaptive testing, the conditions of application and the possibilities of web technologies for its organization. The article presents the results of a study aimed at organizing and conducting adaptive knowledge control as one of the means of implementing an individual learning trajectory. The study presents algorithms for constructing an individual trajectory of adaptive testing for each user. The analysis of web technologies and learning management systems that are currently used in terms of their capabilities for the implementation of the adaptive learning module is carried out. An adaptive testing module has been developed for implementation into the LMS Moodle learning management system, built taking into account the modular organization of the system. The construction of the module is based on a Markov random process with discrete states, continuous and discrete time, which makes it possible to implement a condition for completing testing with the function of viewing detailed statistics of its passage. The problem of developing and implementing an adaptive testing module in an online learning system is considered. The results of experimental work confirming the effectiveness of the implementation of the adaptive testing module are presented. The study of technology by future teachers and its application in practice will contribute to an increasingly widespread implementation in practical activities.

The problem evaluated in this study is related to the optimization of a budget of an industrial enterprise using simulation methods of the production process. Our goal is to offer a universal and straightforward methodology for simulating a production budget at any level of complexity by presenting it in a specific form. The calculation of such production schemes, in most enterprises, is currently done manually, which significantly limits the possibilities for optimization. This article proposes a model based on the Monte Carlo method to automate the budgeting process. The application of this model is described using an example of a typical meat processing enterprise. Approbation of the model showed its high applicability and the ability to transform the process of making management decisions and the potential to increase the profits of the enterprise, which is unattainable using other methods. As a result of the study, we present a methodology for modeling industrial production that can significantly speed up the formation and optimization of an enterprise’s budget. In our demonstration case, the profit increased by over 30 percentage points.

To predict the spread of the new coronavirus infection COVID-19, the critical values of spread indicators have been determined for deciding on the introduction of restrictive measures using the city of Moscow as an example. A model was developed using classical methods of mathematical modeling based on exponential regression, the accuracy of the forecast was estimated, and the shortcomings of mathematical methods for predicting the spread of infection for more than two weeks. As a solution to the problem of the accuracy of long-term forecasts for more than two weeks, two models based on machine learning methods are proposed: a recurrent neural network with two layers of long short-term memory (LSTM) blocks and a 1-D convolutional neural network with a description of the choice of an optimization algorithm. The forecast accuracy of ML models was evaluated in comparison with the exponential regression model and one another using the example of data on the number of COVID-19 cases in the city of Moscow.

The main objective was to study the milk productivity of lactating cows depending on the level of total accumulation of toxic metals in the blood serum. The experimental part of work was carried out on clinically healthy Holstein cows (n=45; lactation stage - 20-40 days after calving) bred in a single farm. The elemental composition of blood serum was studied for 4 chemical elements (Al, Pb, Sn, Sr) by ICP AES and ICP MS. The indicators of milk productivity of cows were evaluated by the yield of milk fat, protein, lactose, dry matter, SOMO and the average daily milk yield. To assess the magnitude of the toxic load on the body of cows, the coefficient of the total toxic load Ktox (µg/ml) was calculated. To calculate the coefficient, the sum of the concentrations of some toxic elements in the blood serum was used: Ktox=Al+Pb+Sn+Sr. Based on the data on the value of the total toxic load coefficient Ktox in blood serum, three groups were formed: group I included cows with a minimum value of the toxic load coefficient (Ktox=0.128 (0.088-0.142) μg/ml relative to the study sample; group II with an average (Ktox=0.229 (0.206-0.276) µg/ml; Group III included cows with the maximum coefficient (Ktox=0.395 (0.306-0.453) µg/ml. Comparative analysis showed that cows with the minimum Ktox value (group I) outperformed individuals from I and group II in terms of milk fat yield by 15.4 (P£0.001) and 19.7% (P£0.001), protein by 5.6 and 12.5% (P£0.05), lactose -10.9 (P£0.01) and 18.4% (P£0.001), dry matter by -7.6 and 12.8% (P£0.05) and average daily milk yield by 9.4 and 17.5% (P£0.01), respectively. Thus, in the course of the study, it was registered that with an increase in the total accumulation of heavy metals in the body of animals, the milk productivity of cows decreases.

Currently, a significant group of industrial facilities can be classified as chemically hazardous facilities (CHFs). To predict the spread of harmful impurities in the programs being developed, Gaussian and Lagrangian models are actively used, on the basis of which the complexes used both in the EMERCOM of Russia and in research organizations are being implemented. These complexes require the introduction of a large amount of information, including the characteristics of the wind field in the distribution of an emergency chemically hazardous substance, which limits their use. In systems, the formation of which is influenced by a large number of different random factors, spatial scaling (similarity) is often found, and one or another parameter can be described using the methods of fractal geometry, which in the past few decades has been actively and successfully applied to the description of various physical objects. The purpose of this study is to analyze the possibility of using the random-addition method for early prediction of the distribution of harmful impurities in the surface air layer during the short-term release of a substance on the surface as a result of an emergency.

The aim of this study was to evaluate the effect of GDF5 (growth differentiation factor 5) gene polymorphism on metabolism of chemical elements and milk production of cows. The studies were performed on Holstein cows (n=100). To determine the single nucleotide polymorphism of GDF5 (T586C in exon 1), blood samples were taken. DNA samples were isolated from whole blood using the IsoGeneLab reagent kit. The thermal cycler "MyCycler" was used for DNA amplification. The elemental composition of blood serum was estimated according to 13 chemical elements (Al, As, Pb, Sn, Sr, Co, Cr, Cu, Fe, I, Mn, Se, Zn) by ICP AES and ICP MS. The frequency of occurrence of polymorphic groups in the GDF5 gene was established, which was for homozygous alleles: TT (n=41), CC (n=10), heterozygous (TS)–(n=49). The analysis of the data obtained showed that blood serum of cows with the CC genotype, relative to animals with the TT and TC genotypes had a greater concentration of I - by 19.3 (P£0.001) and 10.4% (P£0.01), Se - by 18.3 (Р£0.001) and 11.7% (Р£0.01), Zn – by 15.8 (Р£0.01) and 13.2% (Р£0.01), while relatively low concentrations of Al were registered - by 10.8 (Р£0.001) and 21.2% (Р£0.01), Pb - by 16.4 (Р£0.01) and 23.7% (Р£0.001) respectively. In terms of milk fat yield (kg/day), animals with CC genotype advanced over animals with the TT and TC genotypes by 4.8 (P£0.05) and 5.2% (P£0.01), protein by 1.2 and 1.4% (Р£0.05), milk yield - by 4.8 (Р£0.05) and 5.3% (Р£0.01). Thus, we can conclude that the metabolism of chemical elements and the milk productivity of cows are significantly determined by the polymorphism of the GDF5 gene. In order to select cows of the Holstein breed according to the GDF5 gene polymorphism, it is advisable to select homozygous animals for the CC genotype.