Blended Enriched Virtual Model for the Prediction of Students’ Performance Using Probablistic Based Model

Abstract

The online mode of learning is gaining popularity in the education field due to various reasons. The rapid expansion in the practice of web-based technologies have made educators to take advantage of ICT based learning in Higher Education Institutes worldwide in emergency situations. The occurrence of novel COVID-19 pandemic shifts the ongoing physical education system from face-to-face learning to virtual learning worldwide. Virtual learning is not sufficient for the fulfillment of undergraduate students’ learning requirments as they require practical knowledge as well. Therefore, higher education institutes need the integration of blended learning methods for the improvement of students’ learning outcomes in their respective field of studies. Blended learning is the latest trend of implementing online learning strategy along with other e-learning tools. This research paper focuses on the development of a blended virtual model using a probabilistic model i.e. Bayesian network classifier for the prediction of students’ academic performance. The blended enriched virtual model is adapted that includes online and offline learning. Online learning includes online lectures, chat collaborations and online courses. Whereas offline face to face learning includes physical classroom lectures and lab sessions for practical work. The proposed BN model is applied to undergraduate computing students for analysis of learning outcomes of Data Structures and Algorithms subject. According to the findings of proposed BN model that if students properly attend the classroom lectures followed by their lab practical in Face-to-face learning and the proper online learning activities like lectures, chats and online courses, the learning outcomes of the students may be improved and the proposed BN model also reports the accuracy of about 85%.