Abstract: The pilot work of the Giant Panda National Park faces various challenges such as the protection of wildlife under habitat fragmentation and the increasing domestic demand for tourism. Based on the network text analysis method, the Giant Panda National Park is taken as a case to analyze the influencing factors and adaptation behavior of tourists' congestion perception. We collect online texts of the Chengdu Giant Panda Breeding Base, Bifengxia and Wanglang Nature Reserves, which are representative and often carry out recreation activities in the Giant Panda National Park, and by using ROST Content Mining 6 computer analysis software and NVivo 8 qualitative analysis software, combined with content analysis and qualitative subject analysis, we analyzed the crowded perception-related text. The text data covers 1711 comments from 2011 to 2018, for a total of 132354 words. The study found that the individual and travel characteristics, scene characteristics and population density of tourists are three important factors affecting the crowd's perception of congestion. As a feedback behavior to the crowded situation, the adaptation response is a special factor influencing congestion perception and an important part of the comment text. Finally, based on the in-depth analysis of tourists in the Giant Panda National Park, relevant suggestions for handling the problem of crowding in national park type tourism destinations are put forward, to promote the national park to achieve the dual goals of wildlife protection and destination recreation management.