Big Data y Rendimiento Académico en la Educación Superior: Una Revisión Bibliográfica basada en PRISMA

Abstract

Big Data has revolutionized several sectors, including higher education, facilitating the collection and analysis of data to support decision making. Its application in academics has proven to be an important tool to predict student performance, personalize learning and generate effective strategies for reducing university dropout. This study is supported by a review of bibliographic sources that apply Big 1 Data in higher education, under the PRISMA methodology, analyzing recent research obtained from databases such as Taylor & Francis, Scopus, ScienceDirect and IEEE Xplore. Big Data, together with machine learning and data mining, creates predictive models to estimate students' academic performance, identify risk factors and optimize learning strategies. Thanks to this, it is possible to detect key factors that affect academic performance, such as emotional problems, economic situation or lack of personalized attention. Among the main findings, it was identified that approximately 75% of the reviewed improvements studies in report academic performance when online learning platforms are implemented. Furthermore, the use of technologies such as Big Data has contributed to personalizing teaching processes, facilitating learning that is more tailored to students' needs. Despite its advantages, its implementation in higher education faces challenges such as data protection, staff training and the difficulty of adapting institutions to new methodologies. With the right strategies in place, Big Data can transform education, improving learning and increasing student retention through data-driven models.