Architectural Enhancement with Virtual Reality Infused Intelligent Hybrid Learning Approach for Elevated Students Performance-Take English Teaching as an Example

Abstract

Visual Reality (VR) applications and technologies are imparted into teaching sessions to improve students' performance through real-time experiences. VR requires diverse modal data to validate the application's success. Therefore, screened data will be imported for training VR applications for precision-oriented results. This manuscript introduces a Multi-Instance Learning for Data Filtering (MIL-DF) for leveraging students' performance through VR applications. The first step is analyzing teaching accumulated data across different time instances. In this process, the data required for the VR system is determined using the trained teaching model. This enhances the filtering precision for delivering tailored training and visualization sessions. The training adapts the pre-filtered and filtered data from different instances for matching better outcomes. Therefore, the learning model is trained from session outputs, performance scale, and accumulated teaching data. This is tuned by reducing the impact of filtered data across consecutive training instances. The training Iterations are planned for giving rapid data validations after the reduction by the training. This enhances the performance-oriented analysis with less training complexity and achievable performance efficiency.