Consumer recognition of service and product attributes in smartphone rivalries
摘要
This study introduces a consumer-centric analytical framework that leverages latent Dirichlet allocation (LDA) to quantify perceived rivalries among smartphone brands in China. Analyzing 2.8 million BiliBili comments (2018–2024), the model systematically measures brand co-mentions, adjusting for mention order and multi-brand contexts, as proxies for competitive intensity. Findings indicate that consumer attention is gradually shifting from hardware attributes, such as price and performance, toward expectations regarding after-sales support and service integration. By capturing these evolving priorities, the framework enables real-time market sensing for manufacturers and service designers and provides actionable insights into how hardware and service dimensions collectively influence competitive dynamics.