Year: 2025
Type: article
Abstract: The formation of an ensemble of key concepts in relation to texts with highlighted semantic parts (text with a table of contents) is reduced to the stratification process from three procedures - the procedure of text preparation, the procedure of extracting key concepts from semantic parts and dividing the entire text into fragments related to the found key concepts. Quantitative characteristics of the ensemble (the number of words in related fragments) make it possible to solve a number of problems, including determining the predominant content of the text, calculating the parameters of text proximity, identifying concepts of interest to the reader, forming fragments for training neural networks while preserving the author's style, etc. The article briefly describes the formation procedures and provides three examples of using the ensembles of key concepts obtained by stratification. In the first example, the most fully (by the number of words in related fragments) disclosed key concepts in textbooks on the subject of "Project Management" in English, German, French and Russian are determined. The results obtained make it possible, for example, to justify the choice of a specific textbook. In the second example, for ten Russian-language educational and methodological publications on project management, proximity parameters were calculated, including the normalized length of the difference vector, the angle between the ensemble vectors, and the normalized integral characteristic. The results obtained can be used in selecting materials for educational programs and individual courses. In the textbooks participating in the first example, the longest continuous fragments of texts by the number of words, suitable for LLM training, were found.
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
SDG: Quality education
Open Access status: hybrid
OpenAlex ID: W4409514484