Predicting the Class of a Bachelor’s Degree Student Using an Artificial Neural Network

Abstract

The academic status of a student following a degree programme is very useful to the university and the student as well in many ways. Through the degree programme most of the students aim to graduate with a class, but very few students achieve their goal. Therefore, both university administration and students are very concerned about their academic status. Through the proposed Undergraduate’s Class Prediction System, we hope to predict a student’s class at an early stage of their respective degree programme using an Artificial Neural Network (ANN). In this work, a multi-layered feedforward neural network is used to classify the class of a student’s degree into first, second-upper, second-lower or general degree. Using the feedforward algorithm, we were able to achieve the best performance with an accuracy of 76.27%. Through this system the university can identify underperforming students at the early stages and help them with the difficulties they face, can help talented students to finish their degree with a good GPA leading to a better class and will be able to identify the potential dropouts and counsel them with academic guidance.