USING TENSORFLOW/KERAS LIBRARIES FOR TRAINING NEURAL NETWORKS IN THE TRAINING OF BACHELORS IN VOCATIONAL EDUCATION OF THE COMPUTER SPECIALTY

S. V. KOZIBRODA Ternopil Volodymyr Hnatiuk National Pedagogical University https://orcid.org/0000-0003-4218-0671
Yu. P. FRANKO Тернопільський національний педагогічний університет імені Володимира Гнатюка https://orcid.org/0000-0002-1464-1162
D. I. KLUBKO Ternopil Volodymyr Hnatiuk National Pedagogical University https://orcid.org/0009-0001-7063-2818

Keywords: TensorFlow, Keras, neural networks, vocational education, digital technologies, artificial intelligence, machine learning

Abstract

DOI: https://doi.org/10.26565/2074-8922-2025-85-13

Purpose. The purpose of the study is to analyze the potential of using TensorFlow and Keras libraries for training neural networks in the preparation of bachelors in vocational education in computer technologies, as well as to develop a methodology for their effective integration into the educational process. Special attention is given to how the practical application of these libraries contributes to the development of digital competencies and analytical thinking among future engineering educators.

Methods. The study is based on a systematic analysis of scientific publications from 2019 to 2025 in the Scopus, Web of Science, and Google Scholar databases focusing on the use of TensorFlow and Keras in education. A comparative analysis of domestic and international experience in preparing specialists in artificial intelligence is conducted. The method of pedagogical modeling is applied to develop a methodology for teaching students how to create, train, and evaluate neural networks using TensorFlow and Keras libraries.

Results. The study substantiates a methodology for preparing students to work with neural networks based on TensorFlow and Keras libraries. It involves step-by-step mastery of building, training, and evaluating machine learning models, which promotes the development of practical skills in implementing intelligent systems. The proposed methodology is aimed at forming students’ understanding of neural network architectures, optimization principles, and practical applications in their professional activities.

Conclusions. The use of TensorFlow and Keras libraries in the preparation of bachelors in vocational education in computer technologies opens new opportunities for the practical acquisition of artificial intelligence technologies and the development of engineering and pedagogical competencies. Prospects for further research include identifying pedagogical conditions for the effective implementation of the developed methodology, assessing its impact on the quality of professional training, and exploring possibilities for adaptation in other technical educational programs.

In cites Kozibroda S. V., Franko Yu. P., Klubko D. I. (2025). Using TensorFlow/Keras libraries for training neural networks in the training of bachelors in vocational education of the computer specialty. Problems of Engineering Pedagogic Education, (85), 149-162. https://doi.org/10.26565/2074-8922-2025-85-13 (in Ukrainian)

Downloads

Download data is not yet available.

References

1. Zinovieva, O. H. (2024). Overview of software for working with neural networks. Visnyk of Kherson National Technical University, 2(89), 157–161. https://doi.org/10.35546/kntu2078-4481.2024.2.22 (In Ukrainian).
2. Kovalenko, O. E., Kovalenko, D. V., Briukhanova, N. O., Melnichenko, O. O. (2017). A modern approach to determining and developing the main components of the educational program for training specialists in specialty 015 "Vocational education (for specializations)". Problems of Engineering Pedagogic Education, 56–57, 6–18. http://nbuv.gov.ua/UJRN/Pipo_2017_56-57_3 (In Ukrainian).
3. Kryvokhata, A. H., Gevorgyan, A. A. (2020). Review of software tools for working with neural networks. In Priority Directions of Science and Technology Development: Proceedings of the III International Scientific and Practical Conference (November 22-24, 2020) (pp. 444–449). Kyiv: Scientific Publishing Center Sci-conf.com.ua. https://sci-conf.com.ua/wp-content/uploads/2020/11/PRIORITY-DIRECTIONS-OF-SCIENCE-AND-TECHNOLOGY-DEVELOPMENT-22-24.11.20.pdf (In Ukrainian).
4. Liubarets, V., Lyubima, A. (2022). Establishing the competitiveness of future specialist junior bachelor’s in software engineering. Higher Education of Ukraine, 1–2, 38–43. https://doi.org/10.31392/NPU-VOU.2022.1-2(84-85).05 (In Ukrainian).
5. Orap, M., Vovk, V. (2022). The model of the development of the future specialists’ competitiveness: theoretical and practical fundamentals. Psychology: Reality and Perspectives, (19), 80–92. https://doi.org/10.35619/PRAPRV.V1I19.315 (In Ukrainian).
6. Tsidylo, I. M., Kozibroda, S. V. (2017). Model of the system of training future engineers-teachers in the field of computer technologies for the study and application of ontologies of computer systems. Scientific Notes of Rivne State Humanities University: Updating the content, forms and methods of teaching and upbringing in educational institutions, 16(59), 154–158. http://nbuv.gov.ua/UJRN/Ozfm_2017_16_49 (In Ukrainian).
7. Tsidylo, I. M., Kozibroda, S. V. (2018). Systems of computer ontologies as a means to form future engineering teachers’ design competences. Information Technologies and Learning Tools, 63(1), 251–265. https://doi.org/10.33407/itlt.v63i1.1838 (In Ukrainian).
8. Yaremenko, V. S., Tarasenko, M. V. (2019). Comparative analysis of software libraries for text data classification using artificial neural networks. Scientific notes of Taurida National V.I. Vernadsky University, 30(69), 1(3), 214–218. https://doi.org/10.32838/2663-5941/2019.3-1/38 (In Ukrainian).
9. Andreykiv, O., Dolinska, I., Nastasiak, S., Liubchak, M., Kozibroda, S. (2024). Mathematical modeling of the influence of natural factors on the durability of critical structural elements. In Proceedings of the 14th International Conference on Advanced Computer Information Technologies (ACIT) 19-21 Sept. 2024 (pp. 112–115). Ceske Budejovice, Czech Republic. https://doi.org/10.1109/ACIT62333.2024.10712531 (In Ukrainian).
10. Buyak, B., Tsidylo, I., Kozibroda, S., Repskyi, V. (2018). Ontological model of representation of university resources. In ICT in Education, Research, and Industrial Applications: Integration, Harmonization, and Knowledge Transfer (pp. 29–40). https://ceur-ws.org/Vol-2393/paper_228.pdf (In Ukrainian).
11. Chicho, V. T., Sallow, A. B. (2021). A comprehensive survey of deep learning models based on the Keras framework. Journal of Soft Computing and Data Mining, 2(2), 49–62. https://doi.org/10.30880/JSCDM.2021.02.02.005
12. Dragan, P., Joshi, K., Atzei, A., Latek, D. (2023). Keras/TensorFlow in drug design for immunity disorders. International Journal of Molecular Sciences, 24(19), Article 15009. https://doi.org/10.3390/ijms241915009
13. Gupta, D. (2023). Image segmentation in Keras: Implementation of SegNet, FCN, UNet, PSPNet and other models in Keras. Computer Science and Engineering. https://doi.org/10.48550/arXiv.2307.13215
14. Melnyk, A., Pasichnyk, R., Androshchuk, U., Honchar, L., Vatslavskyi, O., Kozibroda, S. (2023). Ontology as a software superstructure to a system for mathematical modeling based on interval data. In Proceedings of the 13th International Conference on ACIT 21-23 Sept. 2023 (pp. 543–548). Wrocław, Poland. https://doi.org/10.1109/ACIT58437.2023.10275570 (In Ukrainian).
15. Melnyk, A., Tymchyshyn, V., Shcherbiak, I., Kozibroda, S., Shevchuk, A., Mudrak, R., Hudratian, S. (2025). AI-powered knowledge assessment: Application of GPT models in educational testing. In 2025 15th International Conference on ACIT 17-19 September (pp. 997–1001). https://doi.org/10.1109/ACIT65614.2025.11185592 (In Ukrainian).
16. Watson, M., Sreepathihalli, D. S., Chollet, F., et al. (2024). KerasCV and KerasNLP: Multi-framework models. Journal of Machine Learning Research, 25(1), 18368 - 18377. https://dl.acm.org/doi/pdf/10.5555/3722577.3722952
17. Yapıcı, M. M., Topaloğlu, N. (2021). Performance comparison of deep learning frameworks. Computers and Informatics, 1(1), 1–11. https://dergipark.org.tr/en/download/article-file/1201877