Improving Measurement of Effectiveness of Blended Learning in Iraqi Education Using SVM

Abstract

E-learning has recently become of great importance, especially after the emergence of the Corona pandemic, but e-learning has many disadvantages. In order to preserve education, some universities have resorted to using blended learning. Currently, the Ministry of Higher Education and Scientific Research in Iraq has adopted e-learning in universities and schools, especially in scientific disciplines that need laboratories and a spatial presence. In this work, we collected a dataset based on 27 features and presented a model utilizing a support vector machine with regression that was enhanced with the KNN method, which identifies factors that have a substantial influence on the model for the type of education, whether blended or traditional.
Furthermore, the dataset used was primarily focused on three key factors: personal information, the impact of e-Learning platforms, and the influence of the Corona virus. The attributes that were measured revealed that social status, computer skills, and the basic platform gave the user enough tools to continue the learning process. The size of the classrooms and laboratories that meet the health safety conditions is the most significant. The goal of our work is to discover a model that predicts how blended learning will be used during and after the coronavirus pandemic and to produce a model with minimal errors.