RMRoman Minailenko

Roman Minailenko

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Roman Minailenko has a total of 12 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Analysis of bOtnet Countermeasures in IoT Systems

    by Roman Minailenko, Pavlo Usik, Liudmyla Polishchuk

    The article analyzes the methods of countering botnets in IoT systems. Today, the Internet of Things has become a popular term to describe scenarios in which Internet connectivity and computing power are spread across a multitude of objects, devices, sensors, etc. The main concept of IoT is the ability to connect all kinds of objects (things) that a person can use in everyday life. These objects (things) must be equipped with built-in sensors or sensors that have the ability to process information coming from the environment, exchange it and perform certain actions depending on the received information. The current lack of standards for the protection of such autonomous networks somewhat slows down the introduction of the Internet of Things into everyday life, so there are numerous vulnerabilities in the rapidly growing field of IoT technologies, which are used all over the world. Information protection and confidentiality is one of the priority components when choosing a certain system. Therefore, without adequate confidence in the security and privacy of user data, the IoT system will be uncompetitive. The Internet of Things can cause huge changes in everyday life, bringing a whole new level of comfort to ordinary users. But if the elements of such a system are not properly protected from unauthorized intervention, with the help of a reliable cryptographic algorithm, they will bring harm instead of good, giving cybercriminals a loophole to undermine information security. Since devices with built-in computers store a lot of information about their owner, including the ability to know their exact location, access to such information can help criminals commit a crime. To date, the level of success of botnet countermeasures depends mainly on organizational and political general conditions. Given that the establishment of cooperation or diplomatic agreements takes time, it can be concluded that the establishment of appropriate relations that legitimize cooperation for joint action is not suitable as an ad hoc scheme to combat current attacks. The situation is aggravated, given that modern botnet infrastructures are not under the responsibility of a single entity. In contrast, distributed peer-to-peer networks operate worldwide, so shutting down local parts (often no more than single computers) is not an effective solution. In general, countermeasures that require close cooperation are generally unfeasible today for both technical and political reasons Experts believe that prosecuting botnet developers is unlikely to have a strong impact on the global threat. Instead, botnets need to be fought on a technical level. Proactive measures should be taken by joint efforts of international security groups together with pro-government structures.

  • Distributed Object Technologies in Information Systems

    by Roman Minailenko, Oksana Konoplitska-Slobodeniuk, Iryna Lysenko

    The article shows that the development of information and telecommunications technologies is at a stage when in a distributed information and telecommunications environment, not only the need for access and exchange of information, but also the implementation of various types of analysis and processing of this information is becoming increasingly important. The widespread introduction of computer technologies into all types of activities, the constant increase in their computing power, the use of computer networks of various scales requires the use of a significant amount of high-performance distributed computing, which in turn leads to a shortage of computing resources when performing various computing processes. An effective way to solve these problems is to use parallel and distributed computing. Currently, issues are being discussed regarding the description of products, technologies and methodologies for creating small and medium-sized information systems. At the same time, technologies and methodologies for building large information systems, which combine a set of local information systems, are practically not considered and discussed. The consequence of this is that even at the design stage, technologies for creating a large information system are chosen that do not meet the requirements. For this reason, the projects being implemented do not receive proper development. The modern level of development of society defines the IT industry as a leading and strategic direction of concentration of intellectual and financial resources. Information and tools for its management (software products of various functional purposes) have acquired the status of information resources (IR). The latter are concentrated within the framework of IS. The unification of resources on the basis of information and communication interaction of IS brings them to the level of corporate information resources. This unification is often called the Unified Information Space (UIS). The implementation of UIS at the level of the state, corporation, enterprise is possible in the case of the creation with subsequent observance of the standard for the interaction between IS and their individual applications In the concept of a single information space, it is necessary to provide that the information resources of the IS, in relation to it, act both as data and as various IS applications. Then, in each of the ISs, part of the data processing methods is implemented as applications accessible from other ISs, in particular, in the case of interaction of two ISs, the first is used by services provided by the second, as a result of which it receives already processed data that can be subjected to further processing by the components of the first IS. This approach corresponds to a distributed, peer-to-peer architecture of interaction. According to this architecture, any applications from different ISs can act as both a client and a server in relation to each other, jointly solving certain tasks. This approach minimizes duplication of applications. The distribution of applications across different ISs makes it possible to achieve an optimal balance of application loading and hardware, which will lead to the effective use of information resources of the systems as a whole. Modern technologies make it possible to create an integrated environment within the IS, and within the framework of the EIP concept, which has the following properties: − does not depend on hardware and system software; − is based on international and industrial standards; − allows you to develop a single information model of representing an enterprise as a set of managed resources and activity flows, configured to implement the rules for managing the collective activities of each specific enterprise; − ensures system extensibility, i.e. simplicity and ease of adding new components to existing IS; − allows you to integrate old applications (legacy applications) into new IS; − assumes natural integration of created IS, which guarantees viability and evolutionary development; − allows you to accumulate, replicate and develop formalized knowledge of specialists; − significantly reduces the total costs of creating IS.

  • Exploring Contemporary Data Analysis Approaches Employing Quantum Algorithms

    by Iryna Lysenko, Roman Mynailenko

    The article examines the integration of modern mathematical methods of data analysis with quantum technologies, forming a new interdisciplinary field – quantum analytics. The relationship between classical methods (PCA, SVM, k-means, Monte Carlo, and Least Squares Method) and their quantum counterparts (Quantum PCA, Quantum Kernels, q-means, QAE, HHL) is analyzed. A comparative table of efficiency and computational complexity is presented, demonstrating the potential for exponential or quadratic speedup when using quantum algorithms (e.g., O(log n) versus O(n³), O(1/ε) versus O(1/ε²)). Particular attention is given to the Variational Quantum Eigensolver (VQE) and Quantum Approximate Optimization Algorithm (QAOA) as practical tools for NISQ-devices. The study highlights the prospects of hybrid quantum-classical models for solving problems in big data analysis, optimization, and forecasting. The obtained results confirm that quantum methods can significantly reduce computation time and provide a foundation for the development of next-generation intelligent systems.

  • Open Access

    RESEARCH OF ARTIFICIAL INTELLIGENCE TOOLS FOR INTELLIGENT DATA ANALYSIS

    by Iryna Lysenko, Roman Minailenko, Serhii Smirnov, Kostiantyn Buravchenko, Nataliia Yakymenko, Oleksii Smirnov, Ірина Лисенко, Сергій Смірнов, Наталія Якименко, Олексій Смірнов

    The article presents a comprehensive study of modern artificial intelligence (AI) tools designed for intelligent data analysis (Data Mining), their computational capabilities, and areas of practical application. The key data analysis algorithms are examined, including machine learning methods, deep learning, clustering, classification, regression modeling, and neural networks. Special attention is given to tools such as Scikit-learn, TensorFlow, RapidMiner, Google AutoML, Power BI, and other platforms and frameworks that provide automation of data processing, analysis, and visualization of large-scale datasets. The study includes a comparative analysis of the advantages and limitations of the most widely used AI tools in terms of model accuracy, computational efficiency, ease of integration, scalability, AutoML support, and the ability to work with unstructured data. Examples of applying AI tools in industry, economics, medicine, the financial sector, and other fields of human activity are provided. The research considers current trends in the development of intelligent data analysis, taking into account the growing role of cloud platforms, automated model-building systems, multimodal AI models, and integration with corporate analytical systems. The results of the study make it possible to determine the most effective approaches and tools for solving applied Data Mining tasks, ensuring the selection of technologies according to requirements for accuracy, performance, and openness. The findings may be used in scientific research, business analytics, digital transformation of enterprises, and the design of intelligent decision-support systems.