Educational data mining involves finding patterns in educational data which can be obtained from various e-learning systems or can be gathered using traditional surveys. In this paper, our focus is to predict the academic performance of a student based on certain attributes of an educational dataset. The attributes can be demographic, behavioural or academic. We propose a method to classify a student’s performance based on a subset of behavioural and academic parameters using feature selection and supervised machine learning algorithms such as logistic regression, decision tree, naïve Bayes classifier and ensemble machine learning algorithms like boosting, bagging, voting and random forest classifier. For selection of the attributes, we plotted various graphs and determined the attributes that were most likely to affect and improve prediction. Experiments with different algorithms show that ensemble machine learning algorithms provide best results with our dataset with an accuracy of up to 75%. This has widespread applications such as assisting students in improving their academic performance, customizing e-learning courses to better suit students’ needs and providing tailor-made solutions for different groups of students.
CITATION STYLE
Gajwani, J., & Chakraborty, P. (2021). Students’ Performance Prediction Using Feature Selection and Supervised Machine Learning Algorithms. In Advances in Intelligent Systems and Computing (Vol. 1165, pp. 347–354). Springer. https://doi.org/10.1007/978-981-15-5113-0_25
Mendeley helps you to discover research relevant for your work.