You can view the full text of this article for free using the link below.

Title: School dropout profiling and prediction approach using machine learning

Authors: Khamisi Kalegele

Addresses: Department of ICT, Open University of Tanzania (OUT), Dar es Salaam, Tanzania

Abstract: Tens of thousands of children drop out of schools in Sub-Saharan Africa while the widely adopted interventions are either reactive in nature or uninformed. The increasing availability of disaggregated data has opened new prospects for proactive interventions using new data technologies in unprecedented ways. However, awareness and skills among wider interest groups including school managers, researchers and developers are at staggering low levels. This paper demonstrates how a machine learning approach can be used to profile students and predict the likelihood of dropping out of school in order to enable proactive interventions and potentially inform youth related policies. Using well known open dataset, supervised and unsupervised profiling approaches are demonstrated, compared and proved to be better performers than a traditional approach. The approaches can be replicated under real production environment using actual data to enable informed interventions and reduce dropouts.

Keywords: school dropout; profiling; prediction; truancy; youth empowerment; machine learning; classification; data-driven; absenteeism; Africa.

DOI: 10.1504/IJITCC.2020.112452

International Journal of Information Technology, Communications and Convergence, 2020 Vol.3 No.4, pp.245 - 258

Received: 23 Dec 2019
Accepted: 08 May 2020

Published online: 04 Jan 2021 *

Full-text access for editors Access for subscribers Free access Comment on this article