A Systematic Review of Machine Learning Techniques for Smart Governance in Higher Education

This study presents a systematic literature review (SLR) of machine learning (ML) applications in higher education governance, addressing the growing need for smart governance frameworks in academic institutions. While ML has seen widespread adoption in sectors like finance and healthcare, its integration into higher education remains fragmented. This review began with an initial review of twenty relevant papers to establish thematic baselines, followed by a systematic screening of over 1,100 records using the PRISMA 2020 methodology. From these, 29 studies were selected for in-depth analysis. The review identifies key trends in ML algorithm usage, evaluation metrics, and application domains—ranging from student performance prediction to strategic planning and faculty assessment. Findings reveal critical gaps in empirical validation, scalability, ethical considerations, and real-world implementation, particularly in low- and middle-income countries. The study highlights emerging opportunities in decentralized governance, MLOps, and explainable AI, and proposes a roadmap for building scalable, context-aware, and ethically grounded ML-driven governance models for academia.