Abstract
For higher education, it is crucial for the institution to forecast early student academic performance in order to give students appropriate facilities for improving their academic records and make it simple for the instructor to detect them. Numerous machine learning approaches, including k-nearest neighbour, support vector machines, Naive Bayes classifiers, logistic regression, etc., have been created to enhance student academic performance. In this paper, we propose a method for predicting student performance using a machine learning neural network. The MATLAB toolbox is used to develop this model, and comparison analysis is carried out utilising a variety of measurement factors, including precision, recall, f-measure, and execution time