Evaluating student levelling based on machine learning model’s performance
- Shatha Ghareeb(corresponding author),
- Abir Jaafar Hussain,
- Dhiya Al-Jumeily,
- Wasiq Khan,
- Rawaa Al-Jumeily,
- Liverpool John Moores University,
- University of Sharjah,
- Al-Maarif University College
Research Output:
Contribution to journal
Article
Research
Peer-reviewOpen access
Abstract
<p>In this paper, a novel application of machine learning algorithms is presented for student levelling. In multicultural countries such as UAE, there are various education curriculums where the sector of private schools and quality assurance is supervising various private schools for many nationalities. As there are various education curriculums in United Arab Emirates, specifically Abu Dhabi, to meet expats’ needs, there are different requirements for registration and success. In addition, there are different age groups for starting education in each curriculum. Every curriculum follows different education methods such as assessment techniques, reassessment rules, and exam boards. Currently, students who transfer to other curriculums are not correctly placed to their appropriate year group as a result of the start and end dates of each academic year as well as due to their date of birth, in which students who are either younger or older for that year group can create gaps in their learning and performance. In addition, pupils’ academic journeys are not stored which create a gap for the schools to track their learning process. In this paper, we propose a computational framework applicable in multicultural countries such as United Arab Emirates in which multi-education systems are implemented. Machine Learning are used to provide the appropriate student’ level aiding schools to provide a smooth transition when assigning students to their year groups and provide levelling and differentiation information of pupils for a smooth transition between one education curriculums to another, in which retrieval of their progress is possible. For classification and discriminant analysis of pupils levelling, three machine learning classifiers are utilised including random forest classifier, Artificial Neural Network, and combined classifiers. The simulation results indicated that the proposed machine learning classifiers generated effective performance in terms of accuracy.</p>
Research Output Information
Output type
Research Output:
Contribution to journal
Article
Research
Peer-reviewOriginal language
EnglishArticle number
3Journal (Volume, Issue Number)
Discover Internet of Things (Volume 2, Issue 1)Publication milestones
- Published - 12/2022
Publication status
Published - 12/2022
Publication IDs
- Scopus: 85140218611
Documents and links
Publication metrics
Metrics
SciVal
FWCI
0.62
PlumX, opens in new tab
Captures
89
Citation count
10
Sustainable Development Goals
- SDG 4 Quality Education
