Research on text analysis of consumer reviews in the catering industry based on the BTM model

The catering industry, which plays a crucial role in China’s economy, is currently facing intense market competition and growth challenges. This study aims to explore the trends in consumer review texts within the Chinese catering sector and their influence on consumption decisions. Utilizing Python technology, we collected a substantial amount of user review data from the Dianping, Meituan, and Ele.me platforms. Employing the Bi-term Topic Model (BTM), we conducted a thorough topic mining and analysis of these comments, identifying five primary topics: taste quality, delivery speed, price concessions, service quality, and hygienic environment. These themes encapsulate the core concerns of consumers when selecting foodservice options. Further analysis indicates that these topics significantly influence consumers' purchasing decisions. The research findings not only offer catering operators targeted strategies for enhancing sales and service quality, but also provide a fresh perspective on understanding consumer behavior and optimizing the customer experience.