​

Abstract: With the rapid development of the Metaverse and virtual reality (VR) technologies, immersive brand experience has become an increasingly important mode of interaction between brands and consumers. Unlike traditional questionnaire-based approaches, which rely on structured and often static data, user-generated content (UGC) provides more authentic, dynamic, and context-rich insights into consumers’ real experiences and emotional responses in actual usage scenarios. Against this backdrop, this study focuses on user reviews of three mainstream VR devices—Meta Quest, HTC Vive, and PICO—collected from Amazon and JD.com, and applies natural language processing (NLP) techniques to conduct a systematic and data-driven analysis of textual information.
Specifically, this study employs multiple text mining techniques, including Chinese word segmentation, word frequency analysis, TF–IDF keyword extraction, sentiment analysis, and word cloud visualization, to mine large-scale user reviews and extract meaningful patterns from unstructured data. Through this analytical framework, the study identifies key dimensions of immersive brand experience, including immersion, interactivity, device performance, and user comfort, as well as the overall emotional tendencies expressed by users in real-world usage contexts.
The findings indicate that user evaluations are predominantly positive, reflecting a generally high level of satisfaction with VR devices and immersive experiences. At the same time, negative feedback reveals functional concerns related to product performance, usability, and physical comfort, suggesting that technological limitations still constrain user experience. Furthermore, comparative analysis across brands demonstrates that different technological characteristics lead to distinct patterns of user perception and evaluation, highlighting the heterogeneous nature of immersive brand experience in virtual environments.
By adopting a UGC-based analytical approach, this study uncovers the semantic structure and emotional mechanisms underlying immersive brand experiences, thereby extending existing research beyond traditional survey-based methods. The study contributes to the literature by providing a data-driven perspective on consumer experience in the Metaverse context and demonstrates the effectiveness of text mining techniques in analyzing large-scale consumer feedback. These findings also offer practical implications for firms seeking to enhance user experience and optimize product design in immersive digital environments.

 

Keywords: Immersive Brand Experience; Virtual Reality (VR); User-Generated Content (UGC); Text Mining; Sentiment Analysis; Natural Language Processing (NLP); Metaverse