Recent Advances in Academic Performance Analysis

Abstract

Academic performance analysis has gained popularity in the past decade. Using various prediction and classification methods, researchers aim to provide clues to help students to improve their performance, and to assist educational institutions to improve quality and make better administrative decisions. This work provides a brief survey of 56 papers related to academic performance prediction, published in 2019 and 2020. Statistics and analysis on the prediction target categories, the target population size, prediction and classification methodologies used, and evaluation metrics are presented. It is found that the most commonly used techniques are decision tree, ensemble methods, and neural networks. Futhermore, these techniques also give the highest accuracy in their target prediction.