Open AccessEducational Data Mining for Student Performance Prediction: A Systematic Literature Review (2015-2021)
Author(s) -
Muhammad Haziq Bin Roslan,
Chwen Jen Chen
Publication year - 2022
Publication title -
international journal of emerging technologies in learning/international journal: emerging technologies in learning
Language(s) - English
Resource type - Journals
eISSN - 1868-8799
pISSN - 1863-0383
DOI - 10.3991/ijet.v17i05.27685
Subject(s) - scopus , computer science , decision tree , decision tree learning , demographics , systematic review , data science , educational data mining , classifier (uml) , data mining , artificial intelligence , medline , political science , sociology , demography , law
This systematic literature review aims to identify the recent research trend, most studied factors, and methods used to predict student academic performance from 2015 to 2021. The PRISMA framework guides the study. The study reviews 58 out of 219 research articles from Lens and Scopus databases. The findings indicate that the research focus of current studies revolves around identifying factors influencing student performance, data mining (DM) algorithms performance, and DM related to e-Learning systems. It also reveals that student academic records and demographics are primary aspects that affect student performance. The most used DM approach is classification and the Decision Tree classifier is the most employed DM algorithm.