Systematic Literature Review on Machine Learning Research in Education

This systematic literature review (SLR) critically examines the published literature since 2012 on the applications of machine learning (ML) in higher education. These applications include student performance classification, retention prediction, experience enhancement, and educational data mining. Machine Learning research has developed at a rapid pace to enhance different aspects of our lives. The educational sciences are at a pivotal point due to the potential influence that ML techniques have and will have on the field. We have aggregated the findings from a range of data sources, offering a thorough and objective perspective on the advancements, challenges, and future directions in this area. The volume of work that needed to be considered in the review clearly indicates the importance and relevance of this application of ML. In this SLR, we collect, analyse, and present the existing literature on ML applications in higher education. We utilise a comprehensive search strategy across multiple data sources, including 11 data sources, identifying all relevant studies published from 2012 to 2024. The inclusion criteria focused on peer-reviewed articles using ML-derived tools in areas such as student performance, retention, and experience. Exclusion criteria were rigorously applied to filter out studies that did not align with the focused domain of higher education and machine learning. The initial results yielded a pool of 16,581 articles, then refined to a substantive selection of 15,788 by removing duplicates, screening for false positives, and conducting quality assessment based on a structured quality control (QC) scoring system. The final set of papers was analysed to extract insights into ML applications' methodologies, effectiveness, and trends in higher education. The insights gained have allowed us to identify key trends, methodologies, and areas for future research. In addition, the surveying process will enable us to highlight publication patterns and critical research areas in the space.