B.S. Ksemidov1, K.K. Abgaryan2

1 SC «SRI PI» (Moscow, Russia)
2 Federal Research Center “Computer Science and Control” of the Russian Academy of Sciences (Moscow, Russia)
1 sokboriswork@yandex.com, 2 kristal83@mail.ru

In an era of rapid technological change, adaptation to the needs of each student is becoming not just desirable, but necessary. Intelligent tutoring systems are being developed for this purpose. These systems are capable of providing personalized content and instant feedback, creating conditions for maximizing the potential of each student.

This project aims to develop a modern, adaptive intelligent system that uses advanced machine learning techniques to significantly automate teacher tasks. Existing tutoring systems suffer from a number of limitations, including the need to manually create educational materials in a given format, which requires significant time. In addition, their focus on assessing knowledge to adjust the curriculum involves creating practical assignments, which is also a time-consuming process. The proposed system represents an innovative approach that seeks to reduce the complexity of the teacher's work through the use of machine learning technologies, in particular language models. The key element is the automation of the creation of video and text educational material, which will significantly reduce the burden on the teacher and increase the effectiveness of course preparation.

In the course of the work, an intelligent tutoring system was developed that includes an ontological domain model for the programming and data analysis course, as well as a model of the education and knowledge control process based on a binary vector overlay model. An ontology is a knowledge base consisting of three components: a concept, relations between concepts, and additional constraints (axioms). The overlay model represents integral assessments of the assimilation of concepts and skills of a subject area on a scale. A vector model is a vector where each element characterizes the degree of assimilation of the corresponding concept in the subject area. Binary is an assessment of the knowledge of a concept in the form of a binary meaning "knows-does not know".

The developed system, due to its flexibility and automation, can significantly improve the quality and efficiency of the educational process, opening up new opportunities for teachers and students. It is planned to further adapt the system for other technical courses and integrate it into the educational process. Further research will be aimed at expanding the functionality and integration with other educational platforms, providing the most comfortable and effective learning environment.

Ksemidov B.S., Abgaryan K.K. Automated preparation of educational material in an intelligent tutoring system. Highly Available Systems. 2025. V. 21. № 2. P. 56−65. DOI: https://doi.org/ 10.18127/j20729472-202502-05 (in Russian)

1. Alesheva L.N. Intellektual`ny`e obuchayushhie sistemy`. Vestnik GUU. 2018. № 1. URL: https://cyberleninka.ru/article/n/intellektualnye-obuchayuschie-sistemy (data obrashheniya: 03.06.2023).
2. Alkhatlan A., Kalita J. Intelligent tutoring systems: A comprehensive historical survey with recent developments. arXiv preprint arXiv:1812.09628. 2018.
3. Mousavinasab E. et al. Intelligent tutoring systems: a systematic review of characteristics, applications, and evaluation methods. Interactive Learning Environments. 2021. T. 29. № 1. S. 142–163.
4. Karpenko A.P., Dobryakov A.A. Model`noe obespechenie avtomatizirovanny`x obuchayushhix sistem. Obzor. Mashinostroenie i komp`yu­terny`e texnologii. 2011. № 7. S. 12.
5. Protege [E`lektronny`j resurs]. URL: https://protege.stanford.edu/ (data obrashheniya: 10.04.2025).
6. TopBraid Composer [E`lektronny`j resurs]. URL: https://topbraidcomposer.org/html/ (data obrashheniya: 10.04.2025).
7. Kitson N.K. et al. A survey of Bayesian Network structure learning. Artificial Intelligence Review. 2023. T. 56. № 8. S. 8721–8814.
8. Briganti G., Scutari M., McNally R.J. A tutorial on bayesian networks for psychopathology researchers. Psychological methods. 2023. T. 28. № 4. S. 947.
9. Kudryavcev V.B. i dr. Modelirovanie processa obucheniya. Intellektual`ny`e sistemy`. 2006. T. 10. № 1-4. S. 189–270.
10. Gao Y. et al. Retrieval-augmented generation for large language models: A survey. arXiv preprint arXiv:2312.10997. 2023. T. 2.
11. Zhao P. et al. Retrieval-augmented generation for ai-generated content: A survey. arXiv preprint arXiv:2402.19473. 2024.
12. Peng B. et al. Graph retrieval-augmented generation: A survey. arXiv preprint arXiv:2408.08921. 2024.