Research on user response in virtual reality environment based on visualization and clustering

Abstract

As the feasibility of virtual reality (VR) therapies is gradually being validated, understanding users' physiological and emotional responses to different VR environments is crucial for optimizing VR experiences and improving user engagement. This study aims to introduce machine learning models to analyze user responses to VR environments and identify user groups with similar physiological and emotional characteristics. This study employs visualization tools and clustering models to explore the role of various factors in shaping user experience. K-Means modeling helps the research team understand users' physiological and emotional responses to VR environments. The Principal Component Analysis (PCA) dimensionality reduction algorithm makes the model visualization easier to understand and implement. This study is conducted on a virtual reality experience dataset and produced many nuanced users experience results. The results of the experiments show that there is an apparent disparity in the experience of VR among users of different age groups. It also shows that different user groups have different levels of suitability for VR. Targeted improvements to VR can provide suggestions for people with different needs. This study has significant implications for analyzing and understanding the emotional responses and preferences of VR users and driving the personalization of VR products.