Predicting Student Academic Success: A Fusion of Machine Learning Technique

In a class of 100 students or more, early prediction of students' final grades will aid teachers in identifying weak students so teachers can modify their teaching methods for raising students' grades. Machine learning algorithms can aid in identifying weak students and directing their attention toward their education. The algorithms Random Forest and Support Vector Machine (SVM) were merged in this work. On the historical data of students in one of the under-graduate courses, applied numerous combinations of algorithms, such as Random Forest and K-Nearest Neighbors, Random Forest,Naive Bayes, etc., and developed a model to forecast students taking the same course in the upcoming term's grades. The model with SVM and Random Forest gives an accuracy of 91 % in predicting grades.