Machine learning in higher education. This item is unavailable.Machine learning in higher education

To design a systematic review protocol with bibliometric analysis on the use of machine learning models to predict academic performance and dropout risk in university students during the period 2024-2026.A systematic review is proposed in accordance with PRISMA 2020 and PRISMA-S, with searches in Scopus, Web of Science, ScienceDirect, SpringerLink, ERIC, PubMed, IEEE Xplore, Dialnet and REDALYC. The protocol defines eligibility criteria, Boolean chains, debugging strategy, extraction matrix, methodological quality assessment and synthesis plan. Records, duplicates, exclusions, included studies and bibliometric distributions should be documented only after running searches in primary databases and verifying metadata, full text and quartiles.The review will examine supervised and assembly algorithms, scholarly and digital data sources, predictive metrics, interpretability, explainability, and early warning systems.