Prediction Model Using Classification Algorithms in Machine Learning for Virtual Learning Environment
Virtual Learning Environment (VLE) is learning instructions via a web-based educational delivery system. Virtual learning has become the need of the hour because of its wide range of courses and regular updates. They are associated with excellent professors and teachers. VLE has high flexibility of scheduling and learning. It is also affordable and caters to non-traditional students. VLE can be applied as a pre-requisite to join MOOCS courses, as most course enrolments withdraw due to incapability of individuals to handle multiple courses. Classification of outcomes is very effective with Machine Learning. This prediction can help to identify the influence of VLE on the performance metrics, which has now become an integral part of education system.
Method: We present here the level of accuracy of classification algorithms in machine learning for VLE on student dataset. We predict if VLE has any effect on student performance. OULAD testimonial data set was considered from Kaggle to illustrate the results. The comparison between KNN, Logistic regression, Random Forest, Naïve Bayes, SVM is made.
Objective: To predict the influence of VLE on student performance. This research is also a check to find “at-Risk and struggling students’ performance and improvement by involvement in virtual learning”. The discussion includes algorithm, flow chart and performance evaluators.
Results: Findings show a very high level of accuracy for all the classification except KNN.
Conclusions: Random Forest and SVM has highest accuracy compared to other algorithms.