Enhanced Predictive Model for Student Engagement in The Moodle E-Learning System

Abstract

Technology integration into education has revolutionized learning paradigms, fostering student engagement and expanding educational opportunities. In response to this transformative trend, higher education institutions worldwide embrace active learning methodologies to cultivate essential lifelong skills among students. Despite the global adoption of Learning Management Systems (LMSs), the landscape of LMS utilization varies significantly across regions. Ethiopia's educational sector, encompassing 45 public universities, showcases a distinct pattern with Blackboard emerging as the predominant platform. However, while LMSs offer personalized e-learning experiences, challenges such as course dropout persist, necessitating proactive intervention strategies. Recognizing the pivotal role of student engagement in mitigating dropout rates within virtual learning environments, this study delves into the analysis of student engagement patterns within the context of Wollo University in Ethiopia. Leveraging a dataset comprising information from 500 students, this research endeavors to develop a predictive model using machine learning algorithms to forecast student engagement levels. Through the deployment of various prediction models, including Artificial Neural Networks (ANN), Support Vector Machines (SVM), Decision Trees (DT), and k-nearest neighbors (KNN), the study meticulously evaluates their performance using established metrics such as recall, precision, accuracy, and F1-Score. The results underscore the efficacy of machine learning algorithms in enhancing classification outcomes, with ANN emerging as the most proficient model, boasting an accuracy rate of 93.10%. This research sheds light on the transformative potential of machine learning in predicting student engagement levels within e-learning environments. By offering insights into effective engagement prediction methodologies, it contributes to the ongoing discourse surrounding educational technology and student retention strategies.