https://repository.hneu.edu.ua/handle/123456789/20137 Sun, 05 Apr 2026 20:02:32 GMT 2026-04-05T20:02:32Z https://repository.hneu.edu.ua/handle/123456789/38819 Назва: Чи надають RAG-системи перевагу? емпіричне дослідження ефективності ChatGPT та NotebookLM при створенні тестів з академічних дисциплін Автори: Venhrina O. Короткий огляд (реферат): In the context of the rapid digitalization of education, automating the creation of assessment materials has become a critical task for educators. The emergence of Large Language Models (LLMs) has opened new perspectives for generating test items; however, the question of selecting the optimal toolkit remains unresolved within the educational community. Specifically, there is a need for empirical verification of the hypothesis regarding whether specialized tools with Retrieval-Augmented Generation (RAG) architecture, such as NotebookLM, provide a significant advantage in quality and reliability over universal chatbots (e.g., ChatGPT) when used by subject teachers lacking complex prompt engineering skills. Objective. The study aims to conduct a comparative analysis of the quality, structural correctness, and cognitive depth of multiple-choice test questions generated using three different AI usage scenarios. Methods. The experimental basis was the educational material of the "Database Organization and Storage" discipline (topics: SQL DDL, DML, Aggregation). A pool of 90 test questions was generated across three scenarios: (A) generation in ChatGPT based solely on the topic title; (B) generation in ChatGPT based on uploaded lecture notes; (C) generation in NotebookLM based on an uploaded source. The quality of the obtained content was evaluated by three independent experts on a 5-point scale according to the following criteria: factual correctness, relevance to the topic, and quality of distractors (incorrect answer options). Additionally, questions were classified according to a simplified Bloom's taxonomy. To verify statistical hypotheses, the non-parametric Kruskal-Wallis H-test and Pearson's test of independence were used. Results. Statistical analysis revealed no significant differences ( ) between the three scenarios for any of the quality criteria. A pronounced "ceiling effect" was recorded for the "relevance" criterion (mean scores 4.8–4.99), indicating the high competence of base models in standard academic topics even without providing context. At the same time, it was found that the NotebookLM tool demonstrated technical instability when generating content in Ukrainian, specifically omitting individual words in question formulations, which led to significantly higher variability in "correctness" scores ( ) compared to the stable results of ChatGPT. Analysis using Bloom's taxonomy confirmed that switching to a RAG system does not automatically increase the cognitive complexity of tasks: the majority of questions in all groups remained at the levels of remembering and understanding. Furthermore, generating plausible distractors remains a weak point for all examined AI tools. Conclusions. The findings refute the assumption of the unconditional advantage of RAG systems for generating tests in standardized disciplines. For an educator, the use of universal chatbots with simple prompts is the most effective method in terms of the time-to-quality ratio. The use of specialized tools (NotebookLM) is advisable primarily for working with unique authorial materials; however, it requires increased attention to the verification of each generated question. Wed, 01 Jan 2025 00:00:00 GMT https://repository.hneu.edu.ua/handle/123456789/38819 2025-01-01T00:00:00Z https://repository.hneu.edu.ua/handle/123456789/38705 Назва: Choice of information tools for learning under modern challenges Автори: Solodovnyk G.; Shapovalova O. Короткий огляд (реферат): The study addresses the problem of selecting information and communication tools for learning under conditions of rapid digital transformation in science and education. It justifies the need for a systematic, multi-criteria decision-making approach and defines a model for rational selection of educational technologies. Based on a comparative analysis of existing methods, the analytic hierarchy process is identified as the most flexible framework for evaluating digital learning tools. A criteria-based decision model and algorithm are developed and implemented as a spreadsheet-based decision support system for selecting online educational platforms. The proposed approach is adaptable, practically applicable across educational institutions, and supports digital transformation while accounting for evolving technologies, formats, and user needs. Wed, 01 Jan 2025 00:00:00 GMT https://repository.hneu.edu.ua/handle/123456789/38705 2025-01-01T00:00:00Z https://repository.hneu.edu.ua/handle/123456789/38277 Назва: On the evolution of neural network Автори: Kuklin V. M. Короткий огляд (реферат): The review considers a brief history of the creation of modern neural networks - language models and the emergence of networks capable of solving problems and supporting scientific activity. It is shown that the main drivers of the evolution of artificial intelligence, in addition to the ability to use knowledge from the Internet and large-scale communication-training with users - have become scaling processes. The features of the application of language models in a variety of human activities and in the communication of billions of people who are trying to solve their particular problems with the help of available models are discussed. Special networks began to actively develop, using many developed technologies for solving problems and forecasting. They returned to the formation of models in the form of computation graphs using numerical methods and continual description. Examples are given and approaches to modeling various dynamic systems in natural science and various fields of human activity are discussed. Wed, 01 Jan 2025 00:00:00 GMT https://repository.hneu.edu.ua/handle/123456789/38277 2025-01-01T00:00:00Z https://repository.hneu.edu.ua/handle/123456789/38138 Назва: On the Use of Artificial Intelligence Systems for Scientific Search Автори: Abramov G.; Kuklin V.; Shapovalova O.; Melnyk I. Короткий огляд (реферат): The possibility of involving artificial intelligence systems in working with researchers in the framework of scientific research is discussed. For this purpose, three stages of artificial intelligence participation in solving these problems are identified. The first is the collection of materials and developments on the identified problem. The second stage of scientific research is the formation of a cycle of auxiliary tasks to clarify the connections between concepts and ideas, identifying qualitative and quantitative dependence. This will require an interactive mode of coordinating the formulations of these problems and clarifying the results. At the third stage, a synthesis of the obtained data and knowledge should be carried out to form a complete scientific theory. In addition to using language models, in particular, at the first stage of data collection, it becomes possible to involve Kolmogorov-Arnold networks to identify the dependence between variables in an analytical-symbolic form at the last stages of scientific research. The procedures for synthesizing an array of data and knowledge require clarification to orient artificial intelligence networks when comparing all knowledge obtained at the second stage of scientific research with known formed Wed, 01 Jan 2025 00:00:00 GMT https://repository.hneu.edu.ua/handle/123456789/38138 2025-01-01T00:00:00Z