MODELING THE SOFTWARE DEVELOPMENT LIFE CYCLE OF INTELLIGENT LEARNING SYSTEMS
DOI:
https://doi.org/10.35546/kntu2078-4481.2025.4.3.9Keywords:
software life cycle modeling, intelligent learning systems, machine learning, artificial intelligence, evaluation metricsAbstract
The article considers the concept, methodology and models of the Software Development Life Cycle (SDLC) of intelligent learning systems (ILS). Existing approaches, key stages of the life cycle, challenges and development directions are analyzed. A hybrid SDLC ILS model is proposed, which combines educational strategies, models, and rules with components of artificial intelligence (AI) and machine learning (ML) in order to preserve the educational nature of this class of systems and at the same time benefit from statistical generalization, automation of management decision-making, adaptability, and the possibility of continuous retraining of the system. The paper presents a formalized approach to the SDLC ILS, which synthesizes approaches to the development of ILS, adaptive hypermedia and modern ML practices. The selected stages of the SDLC ILS are described. Methodological approaches, evaluation metrics and management mechanisms are defined for the stages. A hybrid SDLC ILS model is proposed, which combines educational strategies, models and rules with ML components in order to preserve the educational nature of this class of systems and at the same time benefit from statistical generalization, automation of management decision-making and the possibility of continuous retraining of the system. The SDLC ILS modeling process considered in the work provides a structured approach to the development, implementation and management of ILS. Particular attention is paid to modeling the integration of cognitive and pedagogical theories with systems engineering, which forms a new standard of quality of educational technologies and allows simultaneously to ensure scalability, reliability, pedagogical validity and ethical transparency of learning systems. The proposed hybrid model and step-by-step life cycle serve as a roadmap for researchers and practitioners who seek to implement ILS with a high level of quality and security of automated learning processes.
References
Stone J., Patel R., Ghiasi F., Mittal S., Rahimi S. Navigating MLOps: Insights into Maturity, Lifecycle, Tools, and Careers // 2025 IEEE Conference on Artificial Intelligence (CAI). Santa Clara, CA, USA, 2025. P. 643–650. DOI: 10.1109/CAI64502.2025.00118.
Kreuzberger D., Kühl N., Hirschl S. Machine learning operations (MLOps): Overview, definition, and architecture. arXiv, 2022. DOI: 10.48550/arXiv.2205.02302
De Silva D., Alahakoon D. An artificial intelligence life cycle: From conception to production. Patterns. 2022. Vol. 3. Article No. 100017. DOI: 10.48550/arXiv.2108.13861
Юрченко К. Ю., Сивицький Ю. І. Моделі керування життєвими циклами складових освітніх ресурсів компанії. Телекомунікаційні та інформаційні технології. 2024. № 2 (83). С. 85–93. DOI: 10.31673/2412-4338.20 24.029918.
Koedinger, K. R., D'Mello, S., McLaughlin, E. A., Pardos, Z. A., & Rosé, C. P. (2015). Data mining and education. WIREs Cognitive Science, 6(4), P. 333–353. DOI: 10.1002/wcs.1350
Dutchak M., Kozlenko M., Lazarovych I., Lazarovych N., Pikuliak M., Savka I. Methods and Software Tools for Automated Synthesis of Adaptive Learning Trajectory in Intelligent Online Learning Management Systems / M. Dutchak, M. Kozlenko, I. Lazarovych, N. Lazarovych, M. Pikuliak, I. Savka // Lecture Notes in Networks and Systems. 2021. Vol. 183. P. 206–217. DOI: 10.1007/978-3-030-66840-2_16.
