International Journal of Scientific Research and Engineering Development

Marine ecosystems represent some of the most complex and knowledge-intensive domains in environmental science. Students and professionals pursuing marine education face distinct challenges: the sheer breadth of content spanning oceanography, marine biology, fisheries management, and navigation systems, combined with the practical need to connect theoretical knowledge with real-world underwater conditions. Traditional one-size-fits-all educational platforms are inadequate for this diversity of learners. This paper presents a Self-Adaptive AI Learning System (SAALS-Marine) designed specifically to address these gaps through the application of meta-learning techniques. The proposed system continuously profiles each learner's knowledge state, cognitive pace, and conceptual weaknesses using a Dynamic Learner Knowledge Graph (DLKG). A Model-Agnostic Meta-Learning (MAML) core enables rapid adaptation to new learner profiles with minimal interaction data, while a three-tier mastery classification engine categorizes learners as Beginner, Intermediate, or Advanced. The system integrates real-time adaptive content delivery, spaced repetition scheduling, and an AI-powered marine concept tutor capable of explaining complex phenomena such as tidal hydrodynamics, coral bleaching mechanisms, and fisheries stock assessment. Experimental evaluation on a cohort of 340 marine science students demonstrated a knowledge retention improvement of 38.4% and a curriculum completion rate of 91.7% over conventional e-learning platforms. SAALS-Marine offers a scalable, continuously evolving educational framework applicable across maritime training academies, coastal universities, fisheries institutions, and individual selflearners, with significant potential to elevate marine literacy and workforce readiness in India's vast coastal regions.

Agesta Jenifer.A, Vignesh.S,"A Self-Adaptive AI Learning System for Personalized Marine Education using Meta-Learning Techniques" International Journal of Scientific Research and Engineering Development, V9(3): Page(227-233) May-June 2026. ISSN: 2581-7175. www.ijsred.com. Published by Scientific and Academic Research Publishing.