A Faculty-Centered AI-Based Personalized Learning Framework for Undergraduate and Postgraduate Students in Higher Education

Abstract

Personalized learning is increasingly relevant in higher education because students entering the same course can differ in prior knowledge, topic-wise performance, learning pace, engagement, and preferred learning resources. A common instructional sequence may therefore provide insufficient support for some learners while offering unnecessary repetition to others. Artificial Intelligence (AI), Machine Learning (ML), and learning analytics can support a more responsive approach by identifying learner patterns and recommending resources or activities according to observed needs. This paper proposes a faculty-centered AI-based personalized learning framework for undergraduate (UG) and postgraduate (PG) students. The framework combines academic performance, assessment results, learning behaviour, resource preferences, and learner level to construct an evolving learner profile. A personalization engine uses this profile to recommend suitable learning resources, topic difficulty, practice activities, and learning sequences. A feedback loop updates recommendations after subsequent learning interactions and assessments. In addition, a faculty-support layer presents summarized information about weak topics, progress, and assessment trends so that instructors can provide targeted academic intervention. The paper defines the research gap, framework architecture, methodology, data requirements, evaluation plan, and ethical safeguards. No experimental results are claimed because implementation and empirical validation are proposed as subsequent research stages.