Predicting Students’ Academic Performance Through Supervised Machine Learning

There are many supervised and unsupervised types of machine learning approaches that are used to extract hidden information and relationship between data, which will eventually, helps decision-makers in the future to take proper interventions. The variety of powerful algorithms used in different areas of daily life that includes our educational system as well. This paper introduces students' academic performance prediction model that uses supervised type of machine learning algorithms like support vector machine and logistic regression. The results supported by various experiments using different technologies are compared and it is showed that sequential minimal optimization algorithm outperforms by achieving improved accuracy as compared to logistic regression. And the knowledge found through this research can help educational institutes to predict the future behavior of students so that they can categorize their performance into good or bad. The objective is not just to predict future performance of students but also provide the best technique for finding the most impactful features to work on like teacher's performance, student's motivation that will eventually decrease the student's dropout ratio.