Zusammenfassung:
The burgeoning panda tourism market in China is attracting an increasing number of domestic and international tourists. This study focuses on the Chengdu Research Base of Giant Panda Breeding as a case study and utilizes Latent Dirichlet Allocation (LDA) modeling and topic-based sentiment analysis to conduct text mining on online travel reviews in both English and Chinese languages. LDA modeling was employed to identify topics within online reviews, with a subsequent evaluation of the importance of each topic. Furthermore, topic-based sentiment analysis was conducted to assess the performance of different topics. Through importance-performance analysis, this study interprets the destination image disparities between English and Chinese reviews from a cross-cultural perspective. The research findings validate the effectiveness of destination image analysis methods, providing valuable insights for tailoring distinct destination marketing strategies that target tourists from diverse linguistic backgrounds.