Predicting Student Performance Using Educational Data Mining

Abstract

Data mining methods have been employed successfully in several industries, including education, where they are known as educational data mining methods. Educational data mining aims to extract in-depth knowledge from raw data to build automated systems that could be used in the educational sector. With the advancement of data mining technologies, it is now possible to mine educational data to enhance educational practices. This study, therefore, uses educational data mining techniques to predict the final grades of secondary school students. This study has employed several Machine Learning (ML) algorithms, such as classification trees, regression trees, logistic Regression, and Multiple Regression. In addition, the R programming language was used to develop the prediction models. The dataset used in this study was obtained from two secondary schools in Portugal. According to the findings, classification trees and logistic Regression fared better than regression trees and multiple Regression.