Multi-Criteria Evaluation of AI-Based Adaptive Learning Platforms in Global Higher Education: A Fuzzy AHP Perspective

The effectiveness of adaptive learning platforms in higher education is shaped by multiple interacting factors that involve subjective judgment and uncertainty. This study employs the Fuzzy Analytic Hierarchy Process (Fuzzy AHP) to examine the relative importance of key evaluation criteria and sub-criteria based on the perceptions of higher education users with prior experience in technology-enabled learning systems. Data were collected through a structured survey of 150 respondents using fuzzy pairwise comparisons. The evaluation framework was structured around four main criteria—Technological Factors, Content Quality, User Factors, and Institutional Support—supported by twelve sub-criteria. The findings show a clear and consistent emphasis on learner-oriented and practical considerations. Within the technological dimension, system usability is ranked as the most influential factor, surpassing platform reliability and internet accessibility. For content quality, content relevance is prioritized over interactivity and multimedia features, highlighting the importance of meaningful and well-structured instructional material. Among user-related factors, learner motivation emerges as the dominant determinant, followed by digital literacy, while engagement level carries comparatively less weight. In the institutional support category, technical support is identified as the most critical element, reflecting the need for timely assistance to ensure uninterrupted platform use. Overall, the results indicate that effective adaptive learning platforms are driven primarily by ease of use, high-quality instructional content, and motivated learners, supported by responsive technical services. The study provides a structured evaluation framework that can assist educators, institutions, and system designers in making informed decisions regarding the development and adoption of learning platforms in higher education contexts.