Improving Students’ Retention Using Machine Learning: Impacts and Implications

The following revision has been made.  

1.  Write a new abstract and introduction.

2.  Figures were fine-tuned and research work was included   

Traditional statistical tools and qualitative techniques were employed in the literature to discover and forecast charac teristics/factors that impact student retention the most. Modeling the links between these early available indicators and a student's future status of engineering persistence can be very useful in improving student retention in engineering. For some years, machine learning approaches have been used in education to predict retention and discover factors impacting retention rates, with better outcomes since 2010. This study focuses on different machine learning techniques used in literature for improving students’ retention; we have identified various factors that might affect the students’ retention and employed SVM and Neural Networks for predicting students’ retention rates.