Predicting Online Learners’ Performance Through Ontologies: A Systematic Literature Review

Abstract

This systematic review sheds light on the role of ontologies in predicting achievement among online learners, in order to promote their academic success. In particular, it looks at the available literature on predicting online learners’ performance through ontological machine-learning techniques and, using a systematic approach, identifies the existing methodologies and tools used to forecast students’ performance. In addition, the environment for generating ontologies, as considered by academics in the field, is likewise identified. Based on the inclusion criteria and by adopting PRISMA as a research methodology, seven studies and two systematic reviews were selected. The findings reveal a scarcity of research devoted to ontologies in the prediction of learners’ achievement. However, the research outcomes suggest that building an ontological model to harness machine-learning capabilities could help accurately predict students’ academic performance. The results of this systematic review are useful for higher education institutes and curriculum planners. This is especially pertinent in online learning settings to avoid dropout or failure. Also highlighted in this study are numerous possible directions for future research.

Keywords

• data mining
• decision tree
• education
• ontology
• Semantic Web
• classification algorithm