A Comprehensive Review of Student Performance Prediction using Machine Learning Techniques

Education plays a crucial role in the lives of individuals, shaping their purpose and adding value to their existence. This paper presented student performance prediction models by applying KNN, SVM and Naive Bayes on educational dataset of school taken from Kaggle. The main objective of this paper is to provide early estimation of the students' performance so that teachers can plan their pedagogy to improve the student learning. Most of the prior research conducted on the same dataset involved comparing KNN and SVM and they have employed a restricted set of attributes from this dataset. We have introduced an additional machine learning technique to extend the comparative analysis of results and we have utilized all available attributes. The experimental results show that SVM is performing better than KNN and Naive Bayes.