Two-layer ensemble prediction of students’ performance using learning behavior and domain knowledge

Satrio Adi Priyambada*, Tsuyoshi Usagawa, Mahendrawathi ER

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

2 Citations (Scopus)


The ability to predict students' performance is important not only for the students but also for academic stakeholders in higher education institutes. Predictions can be made by using data stored in an academic information system on students' behavior related to taking courses that are an important part of a higher education institute with a coherent vertical curriculum. A student's course-taking behavior can be used as an indicator of their potential performance by investigating the alignment of their course-taking activities with curriculum guidelines. Domain knowledge is also considered as a variable due to the varying compositions of courses in curriculum guidelines. Past performance also needs to be taken into consideration. The result of the prediction can be used to help academic stakeholders take actions such as intervening to ensuring that students graduate on time. In this paper, we propose a two-layer ensemble learning technique that combines ensemble learning and ensemble-based progressive prediction and it utilizes students' learning behavior data and domain knowledge for current and past performances. The results show that the accuracy of our proposed framework on a real-world student dataset is improved.

Original languageEnglish
Article number100149
JournalComputers and Education: Artificial Intelligence
Publication statusPublished - Jan 2023


  • Educational data mining
  • Ensemble learning
  • Learning behavior
  • Students' performance


Dive into the research topics of 'Two-layer ensemble prediction of students’ performance using learning behavior and domain knowledge'. Together they form a unique fingerprint.

Cite this