Machine Learning Algorithms to Predict Students’ Programming Performance: A comparative Study

Research on students’ academic performance is swiftly greater than ever in the field of education, especially students’ performance in programming is crucial. Predicting the performance of students in programming using machine learning algorithms and comparing them to suggest a best model will bestow benefits to the students and teachers. Thus a study has been carried out to suggest a best model for students’ learning in program by comparing the experiments results of Naïve Bayes and Decision Tree, K-Nearest Neighbor, Support Vector Machine and Random Forest algorithms. Data collection, pre-process and classification process are the sequence of steps for building and comparing the models. Test results indicate that Naïve Bayes confers the best accuracy of 91.02% and SVM algorithm has a high accuracy of 88.77%.