2023

Постійне посилання на фонд

https://ela.kpi.ua/handle/123456789/55680

Переглянути

Перегляд 2023 за Назва

Зараз показуємо 1 - 20 з 42
  • Ескіз
    ДокументВідкритий доступ
    Analisysis of the converter with fourteenzone voltage regulation
    (КПІ ім. Ігоря Сікорського, 2023) Mykhailenko, V.; Svyatnenko, V.; Tsyvinskiy, S.; Voytyuk, V.
    The article focuses on the analysis of electromagnetic processes in the electric circuits with semiconductor switches. The mathematical model of the converter with fourteenzone regulation of output voltage has been developed to analyse electromagnetic processes in semiconductor converters with pulse-width regulation. The graphs representing electromagnetic processes in electric circuits are given.
  • Ескіз
    ДокументВідкритий доступ
    Analysis of the use of distance learning in the process of studying professional disciplines
    (КПІ ім. Ігоря Сікорського, 2023) Marchenko, O.
    Distance learning can help make higher education more accessible to different groups of students, including those who work or live remotely from universities. This learning format requires teachers to rethink and adapt teaching methods to the virtual environment. Research into best practices in distance learning can help improve the quality of education. Evaluating the effectiveness of distance learning in the study of professional disciplines is key to ensuring the quality of education and understanding its impact on student achievement.
  • Ескіз
    ДокументВідкритий доступ
    Application of artificial intelligence models to solve the problem of loss of control over the drone
    (КПІ ім. Ігоря Сікорського, 2023) Akhaladze, A.; Lisovychenko, O.; Bogdanova, N.
    This article considers an approach to solving the problem of losing control over a swarm of drones when the connection with the IoT infrastructure is lost. The use of artificial intelligence models of Azure Cognitive Service was proposed to search and classify the command by the operator when creating a duplicate control system. The designed control system has the possibility of deployment on each drone before the execution of the flight plan, maintenance of performance in conditions of lack of communication with the IoT infrastructure in the presence of only visual contact with the operator.
  • Ескіз
    ДокументВідкритий доступ
    Automated detection of the data product consumers in data mesh
    (КПІ ім. Ігоря Сікорського, 2023) Vlasiuk, Y.; Onyshchenko, V.
    Data product became one of the core principles based on which Data Mesh architecture is defined. Being a main and recommended unit for collaboration between domains in a mesh, data product plays the role of communication contract between the components. Distributed nature and high scale of Data Mesh might lead to significant and uncontrolled growth of data product usage by various consumers. This article analyzes existing approaches and tools for detecting data product consumers in Data Mesh and proposes alternative automated approach. Component diagram and execution algorithm are designed as part of the research and presented in the article.
  • Ескіз
    ДокументВідкритий доступ
    Automation system for finding parking places
    (КПІ ім. Ігоря Сікорського, 2023) Napadii, O.; Tsopa, N.; Batrak, Ye.
    This article explores the potential of an automation system designed to revolutionize the process of finding parking places. Through in-depth research, the article highlights the challenges faced by drivers and the shortcomings of traditional methods. The automation system, backed by advanced algorithms and real-time data, aims to enhance parking accessibility, provide accurate predictions of parking availability, optimize driving routes, and improve the overall parking experience for users.
  • Ескіз
    ДокументВідкритий доступ
    Data recognition in documents and classification algorithm
    (КПІ ім. Ігоря Сікорського, 2023) Palii, V.; Zhurakovska, O.
    The article considers the actual problem of data recognition in documents and their classification using the "Core Vocabulary", which corresponds to the common data model for describing public service. For solving the issue an algorithm is developed that allows to recognize data in documents and classify it, which is very important while transferring from a document-oriented public services model to a data-oriented model. The algorithm is the basis of the algorithmic software of “Information system for data recognition and classification”. An illustrative example is present.
  • Ескіз
    ДокументВідкритий доступ
    Detection of face spoofing attacks on biometric identification systems
    (КПІ ім. Ігоря Сікорського, 2023) Zhuravlov, D.; Polshakova, O.
    The article considers methods for the detection and recognition of spoofing attacks on biometric protection systems using the human face, analyses their qualitative indicators and analyses the approach using convolutional neural networks that would allow to obtain the best HTER for the future protection system. The obtained result allowed to highlight the advantages and disadvantages in the design of an attack detection system in the considered area of application. The proposed algorithm of the spoofing attack detection system based on convolutional neural networks using image depth maps.
  • Ескіз
    ДокументВідкритий доступ
    Graphical shell for constructing user-entered arithmetic functions
    (КПІ ім. Ігоря Сікорського, 2023) Smolij, V.; Smolij, N.; Lisovychenko, O.
    The article has a relevant topic in the scientific and practical aspect development of a graphical shell of a software application for constructing functions of two variables entered by the user. The choice of the programming language and the use of the OpenGL software interface are justified. The quality of the construction of the framework of the function depending on the calculation step was investigated. A technique for calculating function coordinates for the applied software interface is proposed. The scaling of the test function and the use of GLSL geometry shaders to create lighting simulation are analyzed. The purpose of the work is to create a graphical shell of a software application for constructing functions of two variables entered by the user.
  • Ескіз
    ДокументВідкритий доступ
    Image generation techniques using generative adversarial networks
    (КПІ ім. Ігоря Сікорського, 2023) Ivanov, A.; Onyshchenko, V.
    GANs were first described in a year 2014, which is quite recently for algorithms. Although, during its time of existence, lots of various modifications and areas of possible usage were found. One of such area is image generation sphere, in which this algorithm is able to achieve results that, in some cases, do not differ from pictures drawn by a person or photographs of certain objects.
  • Ескіз
    ДокументВідкритий доступ
    Intelligent control of a group of transport underwater robots using a coordinator robot
    (КПІ ім. Ігоря Сікорського, 2023) Tymoshyn, Y.; Shevchenko, M.
    The article addresses the problem of utilizing a group of underwater transport robots in an uncertain environment by implementing a robot-coordinator. It includes the analysis of situations, a management model, and the internal and external environment of the robot group.
  • Ескіз
    ДокументВідкритий доступ
    Learning rate in the reinforcement learning method for unknown location targets searching system
    (КПІ ім. Ігоря Сікорського, 2023) Albrekht, Y.; Pysarenko, A.
    The article explores the dependence of the system learning rate on the number of mutually independent modules in the reinforcement learning method. The study defines an environment with two types of objects that bring points to the final score and uses Deep Q Learning algorithms with 36 input data and 5 possible outcomes to conduct the experiment. The goal is to determine the optimal number of objects for which the use of reinforcement learning will give the best result for the same number of iterations. The research is part of a solution to the problem of creating a drone flock control system to find the position of objects in an unknown area.
  • Ескіз
    ДокументВідкритий доступ
    Lightweight agent-based game ai architecture
    (КПІ ім. Ігоря Сікорського, 2023) Hazin, K.; Stetsenko, І.
    The article is devoted to the research game Artificial Intelligence (AI) and architectural solutions for its development. Game AI is one of the most complicated parts of game development, and it needs good tools to reduce complexity and speed up the development. However, there is a lack of lightweight solutions, which can be easily implemented, providing the desired flexibility and reduced complexity. A comparison of game and academic AI is presented, and it is also explained why standard academic AI techniques can’t be broadly applied to game development. Considerable attention is paid to examine existing alternatives such as GAIA, SOAR and AI.Implant, their advantages and disadvantages. The proposed solution for a lightweight agent-based game АІ architecture is described in detail with examples. In addition, the solution provides a space for improvement and extension, which can be useful for more complicated cases than described in the article.
  • Ескіз
    ДокументВідкритий доступ
    Low-resource text classification using cross-lingual models for bullying detection in the ukrainian language
    (КПІ ім. Ігоря Сікорського, 2023) Oliinyk, V.; Matviichuk, І.
    This paper aims on building bullying detection model for Ukrainian language. Considering absence of labeled datasets for bullying detection and classification in Ukrainian, small Ukrainian dataset (4k samples) was gathered and used for testing models in this research. Taking into account very small number of Ukrainian datasets in general this dataset is publicly available for testing and benchmarking other text classification models. Modern approaches to text class classification in low-resource languages are studied in the paper. We apply zero-shot technique and evaluate performance of modern multilingual, cross-lingual state-of-the-art models and embeddings for text classification in Ukrainian language, including mBERT, XLM-R, LASER and MUSE. Experimental results shows that zero-shot approaches for classification task allow to achieve F1 score of 67-69% for multilingual models trained on English dataset only, having 88-91% test accuracy on English data. We also show that machine translation of English data can be used for estimating model performance in other languages, i.e. only 0-2% difference in test accuracy compared to natural data was received for best models XLM-R and LASER. Zero-shot approach for binary detection task showed even better results 81% compared to 91,59% on original English data. We then enhance the best XLM-R model by training it on our natural Ukrainian dataset and confirm benefits of augmenting low-resource language dataset with machine transla tions from resource-rich English data. Finally, the model for bullying detection in the Ukrainian language is built achieving F1 score of 91,59% with only 12k samples dataset in different languages.
  • Ескіз
    ДокументВідкритий доступ
    Models for analysis of water suitability
    (КПІ ім. Ігоря Сікорського, 2023) Makarchuk, L.; Likhouzova, T.
    The problem of unsuitability of available drinking water for safe consumption is considered. It is proposed to use models built by machine learning methods so that when analyzing water samples it is possible to focus on the main parameters so that limited resources are not directed unnecessarily to less important features. To evaluate the effectiveness of the proposed models, test data that were not used to build the models and several different criteria for evaluating the quality of the models were used.
  • Ескіз
    ДокументВідкритий доступ
    Models for forecasting flight delays
    (КПІ ім. Ігоря Сікорського, 2023) Tarasonok, D.; Oliinyk, Y.; Likhouzova, T.
    The problem of improving the operation of airports and air carriers is considered. It is proposed to use machine learning models and technologies to predict flight delays. Several different quality measures are used to evaluate the effectiveness of the proposed models, which diversely reflect the expediency of using these models in the context of the needs of each task.
  • Ескіз
    ДокументВідкритий доступ
    Multi-class classification of pulmonary diseases using computer tomography images
    (КПІ ім. Ігоря Сікорського, 2023) Smilianets, F.; Finogenov, O.
    This paper examines approaches to classifying pulmonary diseases using neural networks. A modification of an existing neural network architecture for multi-class classification based on CT scans is proposed. The proposed architecture distinguishes between coronavirus pneumonia, non-hospital pneumonia, and healthy lungs. The training procedure of the proposed neural network, final parameters, and classification results are described. Conclusions are drawn regarding the potential applications of the proposed modification.
  • Ескіз
    ДокументВідкритий доступ
    Overview of RPA technologies
    (КПІ ім. Ігоря Сікорського, 2023) Troianovska, A.; Batrak, Ye.; Tsopa, N.
    The article's focus is on RPA (Robotic Process Automation) technologies and how they affect businesses' operational procedures. RPA is a software technology that enables businesses to automate repetitive and boring processes with the use of built-in algorithms created by various RPA platforms and vendors. This article's objective is to examine, organize, and determine the potential futures for RPA deployment in the contemporary business environment. The article goes into great detail on the main advantages of RPA implementation into organization system, such as improving operations' productivity, efficiency, and accuracy; cutting down on job completion time; and minimizing errors. A lot of focus is placed in the article on the examination of RPA implementation's potential futures. It is anticipated that RPA's functionality and capacities will continue increasing over time, and that RPA will be integrated with a variety of cutting-edge technologies, elevating business process automation to new heights. In conclusion, the application of RPA technologies is a big step toward the optimization of business processes in a corporate environment that is changing quickly.
  • Ескіз
    ДокументВідкритий доступ
    Preprocessing of audio data for voice transcription systems
    (КПІ ім. Ігоря Сікорського, 2023) Drahan, M.; Pysarenko, A.
    Voice messages are currently a powerful data collection tool. The aim of the study is to speed up the transcription of audio files. To achieve the goal, it is suggested to use a bandpass filter with a lower passband frequency in the range of 150-200 Hz and an upper passband frequency in the range of 3500-7000 Hz. The success of the system is based on the selection of the filter that optimally speeds up the transcription of audio files.
  • Ескіз
    ДокументВідкритий доступ
    Pro-russian propaganda recognition and analytics system based on text classification model and statistical data processing methods
    (КПІ ім. Ігоря Сікорського, 2023) Bezliudnyi, Y.; Shymkovych, V.; Kravets, P.; Novatsky, A.; Shymkovych, L.
    In this paper a neural network model for classifying the political polarity of text has been developed, along with a database for training the neural network and an analytics system for pro-Russian propaganda. This allows to classify the political polarity of the message source based on its identifier, as well as to construct and display different networks that represent useful insights about popular Twitter hashtags or Telegram channels that related to Russo Ukrainian War. Also, a user interface has been developed that allows users to interact with the system. Developed system will help people with navigation through the information space and avoidance of pro-Russian propaganda.
  • Ескіз
    ДокументВідкритий доступ
    Proactive automatic up-scaling for Kubernetes
    (КПІ ім. Ігоря Сікорського, 2023) Gutman, D.; Syrota, O.
    Container management systems are widely used in cloud computing. The leader of the market is Kubernetes. There no ability to configure a proactive scaling for your service using Kubernetes. The research presents an autoscaling technique based on a proactive approach to scale services in Kubernetes.