Machine Learning Enhanced Stress Assessment and Learning Outcome Analysis using Skin Conductance in Virtual Reality Environments

This proposed system explores the integration of virtual reality (VR) technology and skin conductance measurement to analyze improvements in learning outcomes and stress levels. By leveraging immersive VR environments, participants engage in interactive learning scenarios designed to enhance comprehension and retention. Skin conductance, a physiological indicator of emotional and stress responses, is recorded to assess cognitive load and emotional engagement during the learning process. The study aims to establish correlations between stress levels, learning efficacy, and engagement in VR-based education. It applies machine learning (ML) to analyze the relationship between learning outcomes and stress, measured through skin conductance in VR environments. Data is collected from participants, including stress indicators and learning performance metrics. Preprocessing involves cleaning the data, normalizing values, and generating features like baseline and peak skin conductance. Machine learning models, like regression, classification, and clustering, are trained to predict learning outcomes or categorize stress levels. The models are evaluated based on accuracy, precision and recall metrics. Insights derived from feature importance and model performance help understand the impact of stress on learning. Modern machine learning models utilize skin conductance data and achieve 89.47% recognition accuracy using a support vector machine.