EduMentor-AI: A Hybrid Adaptive Intelligence Framework for Personalized Learning in Higher Education

Personalized learning is increasingly essential in higher education due to variations in student abilities, learning pace, and academic preparedness. This paper presents EduMentor-AI, a hybrid adaptive intelligence model designed to support personalized learning through the integration of machine learning, learning analytics, and intelligent mentoring mechanisms. The proposed framework constructs dynamic learner profiles by continuously analyzing academic performance, engagement patterns, and interaction behavior. Based on these profiles, EduMentor-AI adaptively recommends learning resources, adjusts content difficulty, and delivers timely feedback via a virtual mentoring interface. In addition to learner support, the model provides educators with predictive analytics to identify at-risk students at an early stage and enable data-driven instructional planning. The hybrid architecture combines automated intelligence with human supervision to ensure transparency, fairness, and pedagogical effectiveness. Experimental observations indicate improvements in learner engagement, academic performance, and intervention timeliness when compared with conventional instructional approaches. Furthermore, the system reduces manual workload for instructors while enhancing individualized student support. The results demonstrate that EduMentor-AI offers a scalable and learner-centric framework capable of enhancing teaching and learning processes in higher education environments. By acting as an intelligent virtual mentor, the proposed model contributes toward inclusive education, improved academic outcomes, and sustainable digital transformation in universities.