An artificial neural network student enrollment prediction model

As a result of the uncertainty in the number ofstudents to be enrolled into the University, planning and budgetary problems have emerged in numerous Universities across the world particularly in Nigeria. Offices responsible for enrolling students are left to conjecture the numbers likely to turn up. This much of the time isn't reliable since it results to problems in the allotted spending plan and stressing of resources. This work demonstrated an artificial neural network student prediction system that can be used to mitigate the problems raised earlier. The dataset used for the proposed student enrollment prediction model were collected from a dataset database domicile at the University of California, Los Angeles. The dataset has relevant features (Admit, Gre, Gpa, and Rank) which were used to train, and test the machine learning model. From these data, the objective will be to identify if a student will get admitted into the University or not. The proposed student enrollment prediction model was evaluated based on prediction accuracy, and it achieved a prediction accuracy of 93.5%. Based on the findings of this study, accurate data that enhances student enrollment prediction is very essential for use to train the neural network prediction model if high prediction accuracy is the objective. Therefore, there is a need for better techniques in collection of and, retrieval student enrollment data