Browsing by Author "Dmytriieva, Iryna S."

Now showing 1 - 3 of 3

Comparative Analysis of Activation Functions in U-Net for Binary Water Segmentation using Sentinel-2 Imagery
(CEUR-WS Team, Aachen, Germany, 2025) Kundenko, Pavlo; Hnatushenko, Viktoriia V.; Tsaryk, Vladyslav Yu.; Dmytriieva, Iryna S.
ENG: The study examines how different activation functions influence the performance of a U-Net model applied to binary water-body segmentation in Sentinel-2 imagery. Using an identical training setup for each experiment, six nonlinearities—ReLU, Leaky ReLU, ELU, PReLU, Swish and RReLU—are individually substituted into the network while all other parameters remain fixed. Comparative evaluation on a held-out validation set reveals that Leaky ReLU provides the most balanced trade-off between precision and recall, making it the preferred choice for accurate water-mask generation. PReLU offers a similar but slightly lower performance, whereas ELU excels at capturing additional water pixels at the cost of more false positives. The findings highlight the importance of activation-function selection in remote-sensing segmentation tasks and suggest further exploration of advanced nonlinearities and larger, more diverse datasets to enhance generalization.
Development of a Method for Using Graphical Diagrams for Queries to Large Language Models
(International Scientific Unity, Lisbon, Portugal, 2025) Dmytriieva, Iryna S.; Yushkovskyi, Pavlo
ENG: This paper is devoted to the development of an innovative method that combines the visual structure of graphic diagrams with the power of large language models (LLMs). Based on experiments with GPT-4, Claude 3, and Gemini 1.5 Pro, it has been demonstrated that the use of formalized diagrams as queries increases the coherence and accuracy of responses by 16–25% compared to traditional text instructions. The proposed approach opens up new opportunities for structured human-artificial intelligence interaction in education, science, and analytics.
Relationship between Economic Development, Forest Resources, and Forest Fires: European Context
(LLC “Consulting Publishing Company “Business Perspectives”, Sumy, Ukraine, 2024) Dobrovolska, Olena; Schmidtke, Knut; Hnatushenko, Viktoriia V.; Sytnyk, Svitlana; Dmytriieva, Iryna S.
ENG: Conservation of forest resources is a prerequisite for sustainable development of human society, both in the context of preventing negative climate change and for economic growth. The study aims to establish or refute the co-dependence between the level of forest cover in European countries and the production of gross domestic product. The study object is the socio-economic systems of the national economies of European countries in relation to the totality of forest resources of the continent. Studying the dynamics of forest cover indicators (the share of forests in the total area of the country and forest area per capita), weighted within the internationally recognized regions of Europe, it is confirmed that the level of forest cover of European countries is gradually increasing. The analysis of forest fire area maps identifies three main groups by the level of vulnerability to forest fires: safe (Northern European countries), conditionally safe (Western European countries), and dangerous (Eastern and Southern European countries). Denmark, Finland, France, Norway, and Finland show a direct correlation between the level of forest cover of a country’s territory and gross domestic product. The results of cluster analyses based on the data from 2000, 2010, 2015, and 2020 confirm the existence of a stable cluster of European countries (34 countries) in which there is one type of relationship between the production of gross domestic product and the level of forest cover of the territory.