Enhancing Educational Outcomes (EEO) using Cloud Based Virtual Learning Systems and Data-Driven Management

Abstract

In today's digital world, the use of cloud-based virtual learning systems has become an integral part of many schools and colleges. With all these advancements in the learning process, there is still some disconnect between the collection of learning data and its utilization in making decisions on the part of the institution. To address this challenge, this study suggests a novel predictive analytics approach to integrating learning data obtained from cloud-based LMSs and applying them for predicting and forecasting students' academic success. To conduct our research, we collected information from 150 students studying at different universities along with data from their 15 educators representing various academic disciplines. This data included the students' attendance rates, completed assignments, grades for tests and quizzes, and their level of engagement with LMS. To analyze our datasets, supervised machine learning methods such as Random Forest, Support Vector Machine, and Decision Tree classification models were utilized. After testing our predictive learning model with four popular evaluation metrics, Accuracy, Precision, Recall, and F1 Score, the results indicated 88% accuracy of predictions made by the algorithm. Confusion matrix confirmed the effectiveness of the developed system in the prediction of poor and good student performance.