Student Retention Model via Machine Learning and Predictive analysis

The main of objective of this research work is to present the design and implementation of a machine learning based Student Retention Model for an institution of higher Learning. The Decision Tree, Random forest and Neural Network algorithms are used in the development of the student retention model. the implementation results show that Decision tree, Random Forest and Neural Network achieved an accuracy of 95.62%, 95.37% & 91.97% respectively. Machine learning is also used to extract important features which provide possible reasons that hinder students from registering for Post Graduate degrees in the same institution of higher learning where their undergraduate studies were conducted. Some of the important features observed to affect student's retention in the same institution are: student's sport activity, financial status, number of years a student has been registered at the institution and their average marks. This model will be improved in future to include psycho social features.