Keywords
recommender systems
selective educational components
personalization of learning
educational process
How to Cite
Abstract
The article presents the development of an innovative system for automated selection of selective educational components (VOC) for university students. The system analyzes the learning outcomes and personal preferences of the user to dynamically form a list of VOCs, which is updated when interests change or additional courses are taken. The goal is to optimize an individual learning path, increase motivation and learning efficiency. The algorithm consists of three key stages:
- Content-Based Filtering – analysis of thematic similarity of courses based on their descriptions and modules;
- Collaborative Filtering – comparison of student profiles with similar outcomes to identify common elective course selections;
- User Preferences – consideration of individual interests and goals of the student. The system continuously updates data on performance and changing preferences to ensure the relevance of recommendations.
To verify the effectiveness of the developed system, an experimental study was conducted on a sample of students of various specialties. The results showed an increase in the average level of success and increased satisfaction with the educational process due to personalized recommendations. The integrated application of three methods provided an accuracy of over 85% in predicting the most relevant EQFs. This system implements a hybrid approach to personalized selection of elective educational components (EECs) based on student preferences. The combination of three methods enables the generation of recommendations tailored to the individual experience of the student and aligned with the choices of other users. Unlike traditional approaches that rely on manual processes or static templates, the system provides dynamic ranking of EECs based on selected categories of interest. Its implementation as an information technology platform built on a modern stack (NestJS + PostgreSQL) ensures scalability and integration into real educational environments.
References
[1] Верховна Рада України. (2014, лип. 1). Закон України № 1556-VII, Про вищу освіту. [Електронний ресурс]. Доступно: https://zakon.rada.gov.ua/laws/show/1556-18#Text. Дата звернення: 19 квіт. 2025.
[2] X. Zhang, S. Liu та H. Wang, "Personalized learning path recommendation for e-learning based on knowledge graph and graph convolutional network", International Journal of Software Engineering and Knowledge Engineering, т. 33, № 1, с. 109–131, 2023. [Електронний ресурс]. Доступно: https://doi.org/10.1142/S0218194022500681. Дата звернення: 08 трав. 2025.
[3] Y. H. Chen, N. F. Huang, J. W. Tzeng, C.-A. Lee, Y.-X. Huang та H.-H. Huang, "A Personalized Learning Path Recommender System with LINE Bot in MOOCs Based on LSTM", у Proceedings of the 2022 11th International Conference on Educational and Information Technology (ICEIT), Ченду, Китай, 6–8 січня 2022 р., с. 40–45. [Електронний ресурс]. Доступно: https://doi.org/10.1109/ICEIT54416.2022.9690754. Дата звернення: 08 трав. 2025.
[4] Y. Ma, L. Wang, J. Zhang, F. Liu та Q. Jiang, "A Personalized Learning Path Recommendation Method Incorporating Multi-Algorithm", Applied Sciences, т. 13, № 10, ст. 5946, 2023. [Електронний ресурс]. Доступно: https://doi.org/10.3390/app13105946. Дата звернення: 08 трав. 2025.
[5] H. Ahmadian Yazdi, S. J. S. Mahdavi та H. Ahmadian Yazdi, "Dynamic educational recommender system based on Improved LSTM neural network", Scientific Reports, т. 14, ст. 4381, 2024. [Електронний ресурс]. Доступно: https://doi.org/10.1038/s41598-024-54729-y. Дата звернення: 08 трав. 2025
[6] J. K. Tarus, Z. Niu та A. Yousif, "A hybrid knowledge-based recommender system for e-learning based on ontology and sequential pattern mining", Future Generation Computer Systems, т. 72, с. 37–48, 2017. [Електронний ресурс]. Доступно: https://doi.org/10.1016/j.future.2017.02.049. Дата звернення: 08 трав. 2025.
[7] НДІ ПІТ. АСУ ВНЗ. (б. д.). [Електронний ресурс]. Доступно: https://vuz.osvita.net/wp-content/uploads/2021/09/ASU_VNZ.pdf. Дата звернення: 19 квіт. 2025.
[8] ПОЛІТЕК-СОФТ. Деканат. (б. д.). [Електронний ресурс]. Доступно: http://www.politek-soft.kiev.ua/index.php?do=products&product=deanery34. Дата звернення: 19 квіт. 2025.
[9] U. P. Panovyk, R. R. Panovyk, R. I. Petriv та B. I. Fedyna, "Automated educational process support system with dynamic formation of optional components", Scientific Papers (Ukrainian Acad. Printing), т. 1, № 68, с. 94–106, 2024. [Електронний ресурс]. Доступно: https://sci.ldubgd.edu.ua/jspui/handle/123456789/13954. Дата звернення: 19 квіт. 2025.
[10] Р. Берк, "Hybrid recommender systems: Survey and experiments", User Modeling and User-Adapted Interaction, т. 12, № 4, с. 331–370, 2002. [Електронний ресурс]. Доступно: https://doi.org/10.1023/A:1021240730564. Дата звернення: 01 трав. 2025.
[11] N. Manouselis, H. Drachsler, K. Verbert та E. Duval, "Recommender systems for learning", у Recommender Systems Handbook, F. Ricci, L. Rokach, B. Shapira та P. B. Kantor, Ред. Springer, 2011, с. 387–415. [Електронний ресурс]. Доступно: https://doi.org/10.1007/978-0-387-85820-3_12. Дата звернення: 01 трав. 2025.
[12] “Documentation | NestJS - A progressive Node.js framework”. [Електронний ресурс]. Доступно: https://docs.nestjs.com/. Дата звернення: 19 квіт. 2025.
[13] “Home”. Docker Documentation. [Електронний ресурс]. Доступно: https://docs.docker.com. Дата звернення: 19 квіт. 2025.
[14] Banks та E. Porcello, Learning React: Functional Web Development with React and Redux. Sebastopol, CA, USA: O'Reilly Media, 2017.
REFERENCES (TRANSLATED AND TRANSLITERATED)
[1] Verkhovna Rada of Ukraine. (2014, Jul. 1). Law of Ukraine No. 1556-VII, On Higher Education. [Online]. Available: https://zakon.rada.gov.ua/laws/show/1556-18#Text. Accessed on: Apr. 19, 2025. (in Ukrainian)
[2] X. Zhang, S. Liu, and H. Wang, "Personalized learning path recommendation for e-learning based on knowledge graph and graph convolutional network," International Journal of Software Engineering and Knowledge Engineering, vol. 33, no. 1, pp. 109–131, 2023. doi: https://doi.org/10.1142/S0218194022500681. Accessed on: May 8, 2025. (in English)
[3] Y. H. Chen, N. F. Huang, J. W. Tzeng, C.-A. Lee, Y.-X. Huang, and H.-H. Huang, "A Personalized Learning Path Recommender System with LINE Bot in MOOCs Based on LSTM," in Proceedings of the 2022 11th International Conference on Educational and Information Technology (ICEIT), Chengdu, China, Jan. 6–8, 2022, pp. 40–45. doi: https://doi.org/10.1109/ICEIT54416.2022.9690754. Accessed on: May 8, 2025. (in English)
[4] Y. Ma, L. Wang, J. Zhang, F. Liu, and Q. Jiang, "A Personalized Learning Path Recommendation Method Incorporating Multi-Algorithm," Applied Sciences, vol. 13, no. 10, p. 5946, 2023. doi: https://doi.org/10.3390/app13105946. Accessed on: May 8, 2025. (in English)
[5] H. Ahmadian Yazdi, S. J. S. Mahdavi, and H. Ahmadian Yazdi, "Dynamic educational recommender system based on Improved LSTM neural network," Scientific Reports, vol. 14, p. 4381, 2024. doi: https://doi.org/10.1038/s41598-024-54729-y. Accessed on: May 8, 2025. (in English)
[6] J. K. Tarus, Z. Niu, and A. Yousif, "A hybrid knowledge-based recommender system for e-learning based on ontology and sequential pattern mining," Future Generation Computer Systems, vol. 72, pp. 37–48, 2017. doi: https://doi.org/10.1016/j.future.2017.02.049. Accessed on: May 8, 2025. (in English)
[7] Research Institute of PIT. ASU VNZ. (n.d.). [Online]. Available: https://vuz.osvita.net/wp-content/uploads/2021/09/ASU_VNZ.pdf. Accessed on: Apr. 19, 2025. (in Ukrainian)
[8] POLYTECH-SOFT. Dean's Office. (n.d.). [Online]. Available: http://www.politek-soft.kiev.ua/index.php?do=products&product=deanery34. Accessed on: Apr. 19, 2025. (in Ukrainian)
[9] U. P. Panovyk, R. R. Panovyk, R. I. Petriv, and B. I. Fedyna, "Automated educational process support system with dynamic formation of optional components," Scientific Papers (Ukrainian Acad. Printing), vol. 1, no. 68, pp. 94–106, 2024. [Online]. Available: https://sci.ldubgd.edu.ua/jspui/handle/123456789/13954. Accessed on: Apr. 19, 2025. (in Ukrainian)
[10] R. Burke, "Hybrid recommender systems: Survey and experiments," User Modeling and User-Adapted Interaction, vol. 12, no. 4, pp. 331–370, 2002. doi: https://doi.org/10.1023/A:1021240730564. Accessed on: May 1, 2025. (in English)
[11] N. Manouselis, H. Drachsler, K. Verbert, and E. Duval, "Recommender systems for learning," in Recommender Systems Handbook, F. Ricci, L. Rokach, B. Shapira, and P. B. Kantor, Eds. Springer, 2011, pp. 387–415. doi: https://doi.org/10.1007/978-0-387-85820-3_12. Accessed on: May 1, 2025. (in English)
[12] “Documentation | NestJS - A progressive Node.js framework.” [Online]. Available: https://docs.nestjs.com/. Accessed on: Apr. 19, 2025. (in English)
[13] “Home,” Docker Documentation. [Online]. Available: https://docs.docker.com. Accessed on: Apr. 19, 2025. (in English)
[14] A. Banks and E. Porcello, Learning React: Functional Web Development with React and Redux. Sebastopol, CA, USA: O'Reilly Media, 2017. (in English)
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Copyright (c) 2025 Anatoliy Fedonyuk, Andriy Paliy, Yaroslav Mykhalchuk, Valentyna Yunchyk