Кафедра інформатики (ІНФ)

Постійний URI для цієї колекції

Перегляд

Останні подання

Зараз показано 1 - 20 з 542
  • Публікація
    Розробка мікросервісної архітектури для ефективного розпізнавання інформації з чеків
    (ХНУРЕ, 2024) Іткін, Д. О.
    This research focuses on the development of a microservices architecture tailored for efficient information extraction from receipts. The objective is to enhance the recognition process and streamline the handling of information from receipts using a modular and scalable microservices approach. The proposed architecture aims to optimize the overall efficiency and performance of information retrieval in the context of receipt processing.
  • Публікація
    Розробка веб-сайту освітнього ресурсу
    (ХНУРЕ, 2024) Кульмінський, Я. К.
    Educational websites play a pivotal role in modern learning environments, offering a plethora of benefits to both students and educators alike. Firstly, they provide accessible and convenient access to a wealth of educational resources, including interactive lessons, instructional videos, and supplementary materials. This accessibility breaks down barriers to learning, allowing individuals from diverse backgrounds and locations to engage with high-quality educational content. Moreover, educational websites foster self-paced learning, enabling students to progress through material at their own speed and revisit concepts as needed. This personalized approach caters to varying learning styles and preferences, ultimately enhancing comprehension and retention of information. Additionally, these platforms often incorporate gamification elements, such as quizzes, challenges, and rewards, which make learning engaging and enjoyable.
  • Публікація
    Переваги та недоліки класифікації одягу на відвідувачах по їх зображенням
    (ХНУРЕ, 2024) Кравченко, Д. С.
    This work is devoted to the analysis of the advantages and disadvantages of using computer vision for the classification of objects, in particular, in the clothing of web application users. Let’s consider various options for using this technology in everyday life, paying attention to its potential and limitations. The study will also cover modern technological solutions already implemented in the market and compare their effectiveness with traditional methods. In addition, we will analyze the impact of computer vision on user interaction with online platforms and the possibilities of its future improvement. Highlighting these aspects will help to understand how this technology can affect our daily life realities and contribute to the development of modern Internet services.
  • Публікація
    Виявлення дронів за допомогою комп’ютерного зору як елемент захисту конфіденційної інформації
    (ХНУРЕ, 2024) Шевченко, О. Т.
    This paper addresses the crucial issue of confidential information security in the context of the rapidly evolving technological landscape, focusing on the challenges posed by the unauthorized use of unmanned aerial vehicles (UAVs), or drones. The increased accessibility and sophistication of drones have escalated the risks to privacy and security across various sectors, including military, commercial, state institutions, and private individuals. This situation underscores the urgency of developing reliable drone detection methods.
  • Публікація
    Використання gpt-моделей OpenAI для створення відповідей на основі власного контенту
    (ХНУРЕ, 2024) Широкорад, К. А.
    This research explores the application of OpenAI’s GPT models in education, focusing on retraining them to provide customized responses aligned with instructional material. By leveraging GPT models, the study aims to enhance teaching quality and student engagement by addressing queries and clarifying educational topics. It examines two retraining approaches and suggests the use of the LangChain framework for optimization. Overall, the research demonstrates the practicality of utilizing GPT models for tailored educational support, offering potential advancements in web development and educational tools
  • Публікація
    Аналіз сучасних тенденцій у сфері технологій класифікації зображень
    (ХНУРЕ, 2024) Шкарупа, А. О.
    In the digital world, images play a key role in visual communication. Image recognition and classification supported by artificial intelligence are becoming increasingly important tasks. Efforts are being made to improve algorithms and reduce computational costs. Transfer Learning and TensorFlow contribute to solving classification problems. Methods of interpreting the results help to understand the solutions of the models. The future of classification technologies promises development in the directions of 3D images, virtual reality and real time data processing
  • Публікація
    Дослідження методів автоматизації сурдоперекладу жестових мов
    (ХНУРЕ, 2024) Шовковий, Є. І.
    The development of software for automatic sign language translation plays a crucial role in fostering social inclusion for individuals with hearing impairments. Addressing the challenge of social inclusion for those with hearing disabilities is a pressing issue, necessitating solutions within the framework of advancing IT and legislative measures that safeguard the rights and equal opportunities of individuals with disabilities. The research focuses on automated sign language translation methods utilizing intelligent technologies. The objective of this study is to develop and explore methods for automating sign language, aiming to enhance the overall quality of life for individuals with hearing impairments
  • Публікація
    Генетичні алгоритми для вирішення транспортних задач з використанням технологій штучного інтелекту
    (ХНУРЕ, 2024) Ясько, О. С.
    This work is focused on researching of artificial intelligence integration into the Traveling Salesman Problem (TSP) using Genetic algorithm. As time and precision are the key factors in the area as logistics, the work aims to provide an introduction to a method that can be applied in order to increase effectiveness of evolution algorithms for route planning. A method that only expands the possibilities for genetic operators such as selection, crossover, mutation and replacement which represent the real life evolution process
  • Публікація
    Розробка застосунку для оформлення замовлення та оплати в закладах харчування
    (ХНУРЕ, 2024) Цехмістренко, К. В.
    The text explores the global impact of industrialization and digital transformation, noting improvements in technological processes, production quality, speed, and the reduction of physical strain for workers. To enhance competitiveness, the necessity for businesses to seek and create new solutions is emphasized in response to the ongoing evolution of digitalization. Digital transformation has altered the approach to business model development, unlocking new opportunities and channels for profit creation. The adoption of familiar technologies such as QR codes for advertising, information delivery, and interactive features is discussed in the context of restaurant innovations
  • Публікація
    Розробка застосунку "персональна бібліотека” для каталогізації і управління прочитаними книгами
    (ХНУРЕ, 2024) Уткін, Є. І.
    This work is devoted to the development of the "Personal Library" application for cataloging and managing read books. The developed system consists of a user interface, saving information about books, reading analysis and tracking reading speed and time spent on a book. The task of the system is to create a database of the user’s read books. The user can add his own information, such as title, author, year of publication, number of pages, etc., as well as choose a book rating and write a review. The system, in turn, will store this information. The JavaScript programming language and the React framework were used to implement the system
  • Публікація
    Порівняльний аналіз фреймворків PHP
    (ХНУРЕ, 2024) Ходонович, А. Б.
    The research aims to analyze various PHP frameworks with a focus on Laravel and Symfony, with an overview of their advantages and disadvantages. Laravel is characterized by simplicity and a powerful ORM, while Symfony is characterized by high performance and modularity. The pros and cons analysis show that Symfony is suitable for complex large applications, while Laravel is effective for quickly implementing small projects
  • Публікація
    Аналіз зображення для визначення найбільш контрасного кольору фону тексту
    (ХНУРЕ, 2024) Суровикін, Ю. В.
    This work delves into the development of a fully automated system for analyzing the color distribution of digital images. This approach uses image processing techniques to identify the color that offers the most contrasting background color for superimposed text. By identifying this color, the method aims to significantly enhance the visual accessibility and distinctiveness of text elements, thereby achieving confidence in the full readability of both black and light text, regardless of the image analyzed. Furthermore, this methodology has the potential to be applied in diverse real-world applications, such as creating accessible web content and generating informative image captions for visually impaired individuals
  • Публікація
    Огляд методу глибокої нечіткої кластеризації даних
    (ХНУРЕ, 2024) Танянський, О. С.
    Deep fuzzy clustering is one of the directions in the development of fuzzy clustering methods. This approach combines the advantages of fuzzy logic and deep learning for efficient data separation and analysis. The main idea is to use a deep neural network to represent data in a new feature space that takes into account the complex relationships between features and helps preserve the structural features of the data. Deep fuzzy clustering can solve the problems associated with matching high-dimensional and fuzzy data, which are often encountered in real-world problems.
  • Публікація
    Аналіз продажів для визначення популярних ігор на ринку
    (ХНУРЕ, 2024) Терещенко, О. О.
    Conducting sales analyses in an online video game store allows you to analyze sales in the modern video game market and provides information about the most popular genres among players. These theses present the main methods for analyzing sales data to determine the most popular game genres. The study uses information obtained from transactions made in the digital store and analyzed by various categories and parameters. Trends such as the popularity of destinations for sellers, the degree of competition, and the dynamics of changes over time are analyzed. It also takes into account the influence of such factors as the latest trends in the gaming industry. The results of this analysis can be useful for game developers and publishers to effectively plan and develop strategies for the development of a product or products. It is also important to note the importance of risk identification, so methods for this will also be considered
  • Публікація
    Методи семантичної сегментації зображення та їх порівняння
    (ХНУРЕ, 2024) Ткаченко, Н. О.
    This work is devoted to research, comparison and implementation of semantic image segmentation methods with the aim of developing an effective algorithm for automatic selection and classification of objects in images. The purpose of the study is to determine the optimal method or combination of methods that provide the highest segmentation accuracy and speed of image processing. Convolutional Neural Networks (CNNs) were chosen as the main method, noting their high accuracy, generalizability and efficiency in different settings. The implementation of the selected method is performed using a framework for deep learning. The research results can be useful for various applications in modern image processing systems and intelligent systems
  • Публікація
    Застосування глибокого навчання до виявлення об’єктів
    (ХНУРЕ, 2024) Стрельцов, О. А.
    The paper considers the key aspects of object detection in images in the field of computer vision. It is shown that thanks to the development of deep learning, in particular convolutional neural networks (CNN), modern object detection systems have become extremely effective. Key technologies such as R-CNN, YOLO, SSD used for object detection in images are identified. Important components of training deep learning models, such as data augmentation and transfer learning, are also discussed. Examples of the application of object detection systems in various fields, such as the automotive industry, medicine, security, and cooking, are given
  • Публікація
    Розробка веб-застосунку «соціальна мережа для публікації творчих робіт»
    (ХНУРЕ, 2024) Супрун, А. Є.
    In our modern world, where technology is rapidly developing and virtual social networks fill our daily lives, there is a need for a platform that will bring together creative individuals, allow them to find common interests and inspiration, and publish their work. In this context, it is proposed to create a social network for the publication of creative works, which will use the Java programming language and the Spring framework. They allow to create large and complex systems with high productivity and efficiency. Spring provides a wide set of tools that simplify and accelerate the development process, providing effective management of users and data in the social network.
  • Публікація
    Аналіз існуючих застосунків для розпізнавання кулінарних страв
    (ХНУРЕ, 2024) Подшивалова, О. Є.
    The functionality of applications related to the recognition of products and culinary dishes, as well as their classification by allergens and belonging to a group of certain products that can be limited or excluded from the diet while following diet therapy, is considered and analyzed. There are no Ukrainian analogues of culinary classification applications, no applications with similar functions or information on their development. Given this, it can be argued that this area is quite promising
  • Публікація
    Огляд методу нечіткої сегментації зображень
    (ХНУРЕ, 2024) Полубєхін, А. А.
    Computational intelligence methods are widely used to solve many complex problems, including, of course, traditional: Data Mining and such new directions as Dynamic Data Mining, Data Stream Mining, Big Data Mining, Web Mining, Text Mining, etc. Fuzzy image segmentation is an image processing technique that uses fuzzy sets to assign each pixel or group of pixels in an image the appropriate degree of belonging to a certain class or region. The basic idea is that each pixel has a probability of belonging to different classes or areas, rather than a fixed class, as in classical segmentation methods
  • Публікація
    Порівняння backend та frontend тестування у веб-застосунках
    (ХНУРЕ, 2024) Прокоп’єв, С. А.
    This work is devoted to comparing backend and frontend testing, analysing their strong and weak sides, their use in ensuring the comprehensive test coverage of the product. The focus will be on both manual and automation aspects of testing and how we can make sure that the critical functionality works as expected to provide the best user experience and not harm firm’s reputation and business operations. As an example, the checkout charging process will be discussed as a part of this work