Logo image
Social-profile-based E-learning recommendation model
Dissertation   Open access

Social-profile-based E-learning recommendation model

Xola Sinazo Ntlangula
Doctor of Philosophy (PHD), University of Johannesburg
2025
Handle:
https://hdl.handle.net/10210/520068

Abstract

Learning Management Systems (LMSs) form a critical part of the infrastructure in Higher Education Institutions (HEIs), for managing, organizing, and delivering learning content and activities. Their importance has grown significantly, especially since the COVID-19 pandemic, which accelerated the global shift toward online education. This increasing reliance on LMSs has revealed a need for more adaptive and personalised approaches that can respond to the diverse backgrounds, behaviours, and learning goals of students, while also supporting educators more effectively. Although personalised learning has shown great promise, most LMSs remain course-centred rather than student-centred, focusing on predefined learning objectives instead of individual student characteristics. This limitation is particularly significant in diverse educational contexts such as South Africa, where students’ varied socio-economic and cultural backgrounds demand a more tailored learning experience. Moreover, existing personalisation methods rely heavily on data generated within LMSs, overlooking the broader social dimensions that shape how students learn and engage. Research consistently highlights the critical role of social dimensions in learning–students gain deeper understanding through collaboration, communication, and community. Yet, few systems leverage students’ social profiles to enhance personalization within LMSs. Addressing this gap, this study proposes the Social-Profile-Based E-Learning Recommendation (SPER) Model, which integrates students’ social profile data to generate personalized learning strategies. Guided by the Design Science Research Methodology (DSRM), this study designs, develops, and evaluates the SPER Model to demonstrate how social profile integration can enhance personalisation in LMSs. The findings show that the SPER Model is technically feasible, educationally beneficial, and socially responsive. Its main advantages include a student-centred design, adaptive recommendations, continuous feedback, and enriched learning experiences. The study contributes to theory and practice in two ways: (1) It introduces a structured ontologybased approach to model students’ social profiles, identifying seven key social aspects influencing online learning. (2) It develops a hybrid recommendation model that combines social, contextual, and behavioural data to produce personalized learning strategies...
pdf
Ntlangula_Final_Thesis5.32 MBDownloadView
Open Access

Metrics

1 Record Views

Details

Logo image