Developing A Machine Learning Predictive Model For Learning Analytics In Higher Education
Year: 2024
Type: article
Abstract: This study presents the development of a Predictive Learning Analytics Model (PLAM) utilizing Educational Data Mining (EDM) and machine learning algorithms to enhance Teaching and Learning Outcomes (TLOs) at Jazan University's College of Business Administration (JCBA). The research aimed to detect hotspots and predict academic failure among JCBA students to implement proactive interventions. Data from surveys and e-registers involving demographic and academic information from 212 participants over three academic years (2020-2023) were analyzed. An .arff file was created from the collected data, featuring 19 attributes plus one class, and analyzed using the WEKA tool with various classification algorithms. The main findings indicate that the Random Forest classifier achieved the highest accuracy, significantly predicting student performance and identifying at-risk students. This model provides valuable insights for data-driven educational strategies, demonstrating the effectiveness of machine learning in improving academic outcomes. The choice of Jazan University was motivated by the need to address educational challenges and leverage data analytics to enhance student success.
Source: Nanotechnology Perceptions
Authors: Hussam Mohammed Alamoudi, Adel Bahaddad
Language: English
Subfield: Computer Science Applications
Field: Computer Science
Domain: Physical Sciences
SDG: Quality education
Open Access status: diamond
OpenAlex ID: W4403873431