Leveraging ensemble deep learning framework for student academic performance assessment via extended reality in learning environments

Purpose of the Research Immersive learning involves use of sophisticated technologies such as augmented reality. To build interactive environment with student centered settings that support learning, augmented reality (AR) and virtual reality (VR) will be used in interaction and competency learning. AR, VR, and mixed reality (MR) are all versions of extended reality (XR) which invigorate education by improving its accessibility, attendance, and results. Major Findings With XR, the evaluation of student is made possible through integration of deep learning (DL) performing in an immersive way, that is, by movement, gaze, and making choices, providing knowledge outside the conventional approaches and enabling more accurate forecasts of the results, detection of at-risk students, and individualized learning courses. The current research presents the Ensemble Learning Framework Student Academic Performance Assessment through Extended Reality in Learning Environments. The combination of deep learning-based neural networks to assess the performance of various products accurately is called ELSA-XRLE. Conclusions A weighted majority voting-based ensemble classifier is used in the model, which uses gated recurrent units, sparse autoencoders, and vanilla recurrent neural networks; the atomic orbital search algorithm that is optimized on hyper parameters to enhance convergence and generalization. Experimental results with benchmark datasets show that ELSA-XRLE outperforms existing approaches in terms of various performance metrics, demonstrating its potential to enhance student academic assessments in immersive learning environments.