PREDICTING STUDENTS«¤?? GRADE SCORES USING TRAINING FUNCTIONS OF ARTIFICIAL NEURAL NETWORK

Abstract

The observed poor quality of graduates of some Nigerian Universities in recent times has been traced to non-availability of adequate mechanism. This mechanism is expected to assist the policy maker project into the future performance of students, in order to discover at the early stage, students who have no tendency of doing well in school. This study focuses on the use of artificial neural network (ANN) model for predicting students«¤?? academic performance in a University System, based on the previous datasets. The domain used in the study consists of sixty (60) students in the Department of Computer and Information Science, Tai Solarin University of Education in Ogun State, who have completed four academic sessions from the university. The codes were written and executed using MATLAB format. The students«¤?? CGPA from first year through their third year were used as the inputs to train the ANN models constructed using nntool and the Final Grades (CGPA) served as a target output. The output predicted by the networks is expressed in-line with the current grading system of the case study. CGPA values simulated by the network are compared with the actual final CGPA to determine the efficacy of each of the three feed-forward neural networks used. Test data evaluations showed that the ANN model is able to predict correctly, the final grade of students with 91.7% accuracy.ª¤?