Machine Learning-based Model for Prediction of Student’s Performance in Higher Education
Atul Garg, Umesh Kumar Lilhore, Pinaki Ghosh, Devendra Prasad, Sarita Simaiya
Abstract
During the pandemic time, most students are learning in online mode without any physical interaction with a trainer. In this pandemic time, in the absence of physical interaction with students, it became very difficult to predict the performance of students. It's important in particular to support high-risk learners and ensure his\her retention, and perhaps to provide outstanding teaching materials and experiences, and also to improve the institution's rating and brand. This research article presents a machine learning-based model for predicting students' performance in higher education. The work also looks at the possibilities of utilizing visualizations & classification techniques to find significant factors in a small number of features that are used to build a predictive model. The research study analysis revealed that SVM (support vector machine), K*, random forest, and Naive Bayes techniques effectively train limited samples and generate appropriate prediction performance based on various parameters, i.e. precision, recall, F-measure.