Кафедра електронних обчислювальних машин (ЕОМ)

Постійний URI для цієї колекції

Перегляд

Останні подання

Зараз показано 1 - 20 з 660
  • Публікація
    Порівняльний аналіз комерційних обчислювальних систем
    (ХНУРЕ, 2024) Мавринський, О. Д.
    This paper focuses on the relevant issue of maintaining commercial data systems. Two solutions (Office 365 ecosystem and Salesforce) were analyzed. A comparative analysis was carried out, during which several flaws were identified. In the first case – complex and overwhelming process of configuration while only small part of the ecosystem's tools, services and capabilities are used, in the second – environment that is very depended on the network quality and lack of option in case of unstable connection. A system of microservices within the local networks was proposed which addresses these flaws.
  • Публікація
    Прикладні застосування систем комп’ютерного зору
    (ХНУРЕ, 2024) Ціпковський, В. О.
    This work is devoted to assessing the applied applications of computer vision systems in different spheres of human’s life. The exploration of the diverse ways in which the computer vision technology is utilized in practical contexts across the various fields was made. Through the analysis of real-world scenarios, this work delves into how computer vision systems are employed in sectors such as healthcare, manufacturing, transportation, surveillance, and entertainment. By examining specific use cases and implementations, research gains insights into the effectiveness, challenges, and potential advancements of these systems. Overall, investigating the applied applications of computer vision systems offers valuable perspectives on their impact on industry and human experiences.
  • Публікація
    Перспективи моделей навчання з підкріпленням у задачах підтримки прийняття інвестиційних рішень
    (ХНУРЕ, 2024) Горенський, Г. Г.
    The investigation aims to evaluate the effectiveness of machine learning algorithms in investment decision-making by examining reinforcement learning models within the investment decision support system. The research includes an analysis of the PPO, IMPALA, and A2C models. The Gym environment was used to simulate automated trading, and it was found that PPO was the most effective in maximizing portfolio profitability. The findings from 30 experiments were compared using Welch's t-test to ensure objectivity and robustness.
  • Публікація
    Розробка голосового асистента на базі мікроконтроллера ESP32
    (ХНУРЕ, 2024) Михайлов, Ю. О.
    This development is devoted to the creation of a personal voice assistant based on the ESP32 microcontroller. Various technologies were used to solve the tasks, including processing and reproduction of analog signals, saving files on flash memory, interacting with external services via the HTTP protocol. The development has significant potential for improvement and expansion of functionality, including taking advantage of the dual-core microcontroller for multitasking, optimization of processing and data transmission to external services, smart home control, voice wake-up capability, etc.
  • Публікація
    Інструменти штучного інтелекту для створення контенту веб-ресурсів
    (ХНУРЕ, 2024) Хрустальов, Є. К.
    The work examines the possibility of using artificial intelligence algorithms in creating and analyzing content on web platforms. Machine learning methods and natural language processing technologies are discussed, which can improve the quality of content, personalize it to the needs of the audience, and ensure the safety of the online environment. The advantages and problems of using artificial intelligence algorithms in content marketing and information management on web resources are considered.
  • Публікація
    Прикладні застосування систем комп'ютерного зору
    (ХНУРЕ, 2024) Ціпковський, В. О.
    This work is devoted to assessing the applied applications of computer vision systems in different spheres of human’s life. The exploration of the diverse ways in which the computer vision technology is utilized in practical contexts across the various fields was made. Through the analysis of real-world scenarios, this work delves into how computer vision systems are employed in sectors such as healthcare, manufacturing, transportation, surveillance, and entertainment. By examining specific use cases and implementations, research gains insights into the effectiveness, challenges, and potential advancements of these systems. Overall, investigating the applied applications of computer vision systems offers valuable perspectives on their impact on industry and human experiences.
  • Публікація
    Алгоритм організації віддаленого доступу до захищеного сегменту корпоративних комп’ютерних мереж
    (ХНУРЕ, 2024) Чепурна, І. С.
    The article considers the application of remote client access in corporate networks using VPN tunneling based on the Proxmox virtual environment. The problem of secure connection of remote users with the internal network of the enterprise through the use of virtualization and VPN technologies is investigated. Emphasis is placed on analyzing the effectiveness of using Proxmox virtualization technology to create secure connections and ensure secure access of remote users to corporate resources. The advantages of using Proxmox-based VPN tunneling in the context of ensuring confidentiality, integrity and availability of data in corporate networks are discussed.
  • Публікація
    Дослідження процесу міграції віртуальної машини
    (ХНУРЕ, 2024) Радченко, В. О.; Міхаль, О. П.
    The study is devoted to the process of migration. Namely, the transfer of a virtualized guest system from one node to another. Migration has been proven to be a key aspect of virtualization, as software is hardware independent at this level. Migration can be performed in offline or connected mode. In the process of migration, the memory of the guest system is transferred to the target node; while the guest's file system will be stored in the shared storage. There are two types of migration without stopping the virtual machine and with stopping the virtual machine.
  • Публікація
    Сценарій побудови захищеного сегменту мережі між віртуальними офісами
    (ХНУРЕ, 2024) Свергун, В. А.
    The article discusses scenarios for building corporate computer networks that take into account the requirements of secure remote access and fault tolerance by creating reliable virtual communication channels. Provision of secure access to nodes of corporate networks is implemented on the basis of hardware and software tools. The article considers the scenario of building a protected network segment between virtual offices according to the site-to-site scheme, where the Mikrotik CHR software is used as the end nodes.
  • Публікація
    Особливості реалізації спільної фільтрації в Music Recommendation System
    (ХНУРЕ, 2024) Сергородцев, І. Д.
    This work is devoted to finding solutions for effective search in music streaming services. The application of Music Recommendation Systems allows the user to learn about new artists, the release of new albums, songs, musical compositions. Two scenarios for the implementation of the joint filtering method were studied. The first scenario – searching for users based on similar music preferences. The second scenario – the search of users is carried out by different (individual) musical preferences. A class diagram scheme is proposed for implementing collaborative filtering in music recommendation systems.
  • Публікація
    Веб-технологій в управлінні проєктами: бібліографічна карта дослідження
    (ХНУРЕ, 2024) Снігур, А. Р.
    In this work, a bibliometric analysis of trends in web technologies in the field of project management was carried out in order to identify key problems and directions of research. The study analyzed various metrics, such as the annual number of publications and the most used keywords. The following keywords were most relevant: knowledge management, agent, teaching, world, community, information technology. The relevance of these keywords indicates that the topics of training, knowledge management, and information technologies were of great interest to the authors during the research. Thus, on the basis of the obtained results, it is possible to draw a conclusion about the potential directions of further research in this area.
  • Публікація
    Аналіз підходів у виявленні аномалій в зображеннях оптичного моніторингу
    (ХНУРЕ, 2024) Мельніченко, Ф. О.
    In today's world, image processing and analysis from optical monitoring sources is an important area of research due to its wide range of applications. Analyzing such images poses a major challenge for researchers and scientists due to the high variability, low resolution, and large volume of satellite imagery data. One of the fundamental tasks of image analysis from optical monitoring sources is anomaly detection. Anomaly detection in optical monitoring images (satellites, drones, unmanned aerial vehicles) is an important area of remote sensing and geospatial analysis that uses advanced algorithms and machine learning models to detect unusual patterns or changes in data that deviate from the norm. This capability is essential for a wide range of applications, from environmental monitoring and urban planning to military defense, reconnaissance and disaster response. In this article, we will look at common anomaly detection methods and technologies.
  • Публікація
    Огляд технології балансування навантаження у міксованих VPN-ланцюгах
    (ХНУРЕ, 2024) Міхнов, Є. Д.
    In this paper, the most common load balancing technologies in mixed VPN chains are considered. A critical analysis of these technologies has been conducted, including scenarios of their use, potential drawbacks, and implementation challenges. This publication serves as a review and is intended to summarize known methods
  • Публікація
    Оцінка ефективності нейромережевої системи для категоризації текстових документів
    (ХНУРЕ, 2024) Рибалов, О. О.
    This paper presents an evaluation of the DistilBERT model’s effectiveness for categorizing Ukrainian text documents. DistilBERT, a streamlined version of BERT, aims to retain the original's performance with reduced size and increased speed. This study focuses on the model's application for classifying texts into legal and non-legal categories using publicly available data, including court decisions and social media posts. The training encompassed several epochs, enhancing the model's adaptation to data peculiarities. The results, including high accuracy and precision metrics, affirm DistilBERT’s efficacy in this context. This research highlights the potential of neural network systems for automating the processing and categorization of Ukrainian texts in various fields.
  • Публікація
    Система визначення ризиків розвитку ПТСР під час дистанційного навчання з використанням згорткових нейронних мереж
    (ХНУРЕ, 2024) Заброда, І. С.
    This article considers the relevance and problematics of diagnosing PTSD in students who study with the help of distance learning in wartime. The experience of the state institution "Institute of Eye Diseases and Tissue Therapy named after V.P. Filatov of the National Academy of Medical Sciences of Ukraine" in diagnosing PTSD by means of eye movement is considered.
  • Публікація
    Web-застосунок для розрахунку маршруту електромобілів
    (ХНУРЕ, 2024) Зюнд, Б. В.
    This work is devoted to the development of a web application for finding and routing to charging stations for electric vehicles. Problems that drivers of electric cars may face were considered. The relevance of the program and its necessity in the modern world, possible functionality, such as wayfinding considering charging stations and preferences of different types of charging stations, are considered. Technologies that will be used were considered. Algorithms for finding the shortest path were considered. Analogues of this program and their shortcomings are considered.
  • Публікація
    Розробка програмного забезпечення для розміщення вакансій
    (ХНУРЕ, 2024) Кузнєцов, Д. О.
    This work was created with the aim of developing an up-to-date workable model related to the placement of a finished product that will be competitive in the labor market and will be relevant for a wide range of clients. The application is a web development for distributing content related to resumes, vacancies, and projects. It is completely competitive and implements all the necessary functionality. The program is capable of storing, searching and retrieving the necessary data for the client, such as information about the client, projects, vacancies, all relevant data, and is also capable of validating them.
  • Публікація
    Механізми пріоритезації потоків мультимедійних даних у комп'ютерних мережах
    (ХНУРЕ, 2024) Герасимчук, Д. В.
    This paper provides an overview of the mechanisms and ways to implement the transmission of multimedia content in global computer networks. The use of combined solutions based on the use of routing, resource reservation, marking and classification of data packets in the QoS architecture helps improve methods for prioritizing inelastic data flows in computer networks.
  • Публікація
    Концепт архітектури web-застосунку для пошуку та прослуховування музичних композицій
    (ХНУРЕ, 2024) Жук, М. В.; Сергородцев, І. Д.
    The work examines the peculiarities of the development of web applications for searching and listening to musical compositions. A scheme of the three-level architecture of the web application is proposed. The first level is the user interface (using React and Redux technologies). The second level is business logic (using the Nest.js framework). The third level is data access (using MongoDB databases in JSON format).
  • Публікація
    Особливості обчислення у туманих обчислень у сучасних IoT системах
    (ХНУРЕ, 2024) Гунько, М. А.; Фролов, Д. Є.
    In this article will be explored the evolution of computational models. Despite the success of cloud computing, challenges such as high latency and lack of location information persist, leading to a push for a new decentralized approach. This next paradigm, termed fog computing, extends the cloud model by bringing computing and storage capabilities closer to data-generating IoT devices at the edge of the network. While fog computing holds promise for various industries, including healthcare and augmented reality, it also poses challenges such as resource provisioning, security, energy minimization, and standardization, necessitating further research and development efforts