Перегляд за Автор "Kuzmin, A."

Зараз показуємо 1 - 3 з 3
  • Ескіз
    Документ
    Analysis of artificial intelligence based systems for automated generation of digital content
    (Хмельницький національний університет, 2024) Pavlova, O.; Kuzmin, A.
    This paper is aimed at the examination of contemporary challenges related to the integration of generative models API of artificial intelligence (AI) into a unified information system to facilitate the automated generation of digital content. In the context of rapid advancements in AI technologies and the increasing demand for diverse and personalized digital content, the integration of API-based generative models emerges as a crucial driver for progress in this field. The research findings underscore the significance of incorporating API-based generative AI models into a unified system, marking a significant step towards automating the process of digital content creation to meet modern market demands. By streamlining content generation workflows, such integration holds promise for enhancing efficiency and scalability while fostering creativity and innovation. Furthermore, the integration of generative AI models into a unified system presents opportunities for the development of personalized and innovative solutions tailored to the needs and preferences of end-users. This not only enhances user experiences but also enables the creation of content that resonates more effectively with target audiences across various domains. The findings gleaned from our research underscore the importance of integration of API-based generative AI models into a unified framework, representing a monumental stride toward the automation of digital content creation that caters to the exigencies of today's market dynamics. By streamlining content generation workflows and alleviating manual intervention, such integration holds immense promise in enhancing operational efficiency, scalability, and adaptability, while simultaneously nurturing a fertile ground for creativity and innovation to flourish. The further efforts of our research team are committed to the practical implementation of this concept and the exploration of its applicability across diverse domains. By continuing to refine and expand upon this integration, we aim to unlock new possibilities for automated content generation and drive further innovation in the digital content creation landscape.
  • Ескіз
    Документ
    Approaches of building a real-world object detector data source
    (Хмельницький національний університет, 2023) Pavlova, O.; Bashta, A.; Kuzmin, A.
    In our constantly developing world virtual, augmented, and mixed reality technologies are becoming integral parts of our daily lives. In the current stage of Information Technology field development, technologies of virtual, augmented and mixed reality can be seen in almost all areas of human life. Nowadays AR is used in Marketing and Advertising, Education, Medicine, Automotive, Game Development, Navigation and other areas of our everyday life. Therefore, object detection is a crucial task in computer vision and AI applications, enabling machines to identify and locate objects within images or video frames. The accuracy and performance of an object detector heavily rely on the quality and diversity of the training data. This paper is aimed at finding the approaches of building a real-world object detector data source to be able to create a model for detecting a sport games surfaces using the Action & Vision App. During this research several structured approaches of building an object detector data source have been built, drawing inspiration from Apple's Create ML documentation on the topic. Additionally, real-world applications available on both the App Store and Google Play that leverage object detection technology were showcased and analyzed. In the course of study a dataset of objects has been collected and then utilized to build a robust detection model, tailored to function seamlessly with Vision and Core ML frameworks on iOS devices. The trained object detection model, informed by the diverse dataset and robust training process, is employed to identify and outline tables and rectangles in each frame of the video stream. The model and the proposed approaches will be further applied to develop the method of object detection in the real world and create a mobile application for sport games simulation, that would help players to practice their skills out of the training field.
  • Ескіз
    Документ
    Information system for earth’s surface temperature forecasting using machine learning technologies
    (Хмельницький національний університет, 2024) Hovorushchenko, T.; Alekseiko, V.; Shvaiko , V.; Ilchyshyna, J.; Kuzmin, A.
    Temperature forecasting is a topical issue in many areas of human life. In particular, climate change directly affects agriculture, energy, infrastructure, health care, logistics, and tourism. Anticipating future changes allows you to better prepare for challenges and minimize risks. The paper presents an information system for forecasting the temperature of the Earth’s surface using machine learning technologies. The forecast is formed by a model adapted to the region, by learning on the basis of historical data and tracking the most inherent patterns. The selection and training of the model was carried out on the basis of the analysis of the characteristics of climatic zones, according to the Köppen classification. A comparison of the performance of models for forecasting the average monthly temperatures of the earth’s surface in different climatic zones was carried out. The analysis of scientific publications confirmed the relevance of the chosen research topic. Modern approaches to forecasting climatic indicators are considered. Methods and approaches to temperature forecasting, their advantages and disadvantages are analyzed. The peculiarities of the application of machine learning methods for temperature forecasting are considered, and the criteria for choosing the most accurate and least energy-consuming methods are determined. The research results made it possible to identify machine learning methods that best adapt to temperature patterns and allow accurate short-term forecasting. An approach for long-term forecasting using recurrent neural networks is proposed. An information system has been developed for forecasting future temperatures depending on the climatic features of the studied territories based on the proposed methods. A concept for further research for the development and improvement of the developed information system has been formed.