A Comparative Analysis of Students’ Academic Performance using Prediction Algorithms Based on Their Time Spent on Extra-Curricular Activities

Abstract

Machine Learning (ML) is being used to predict trends, hidden patterns in datasets. As the education forms the basic building block of any nation’s prosperity, character and evolution, ML techniques are being exploited for prediction of academic performance of students. This research is about prediction of academic performance of the students based on how much time they spend on extra-curricular activities using various ML prediction algorithms. Also, comparison of the prediction results based on accuracy and F1 Score of these different Machine learning models is made so as to choose the best model for prediction. The research, being at basic level is intended to pave way for further more intricate/ specific and detailed research using more ML models such as Naïve Bayes (NB), Sequential minimal optimization (SMO), Multi-Layer Perceptron (MLP), J48, Random Forest, Random Tree etc for prediction of Academic performance forecast.