Taking Shanghai as an example, this study obtained the online travel notes data from Xiaohongshu and Qunar in the past 10 years to construct the Shanghai tourist flow network (STFN) and used the methods of change point detection (CPD) and complex network analysis (CNA) to reveal the spatial structure characteristics of Shanghai tourism flow and the dynamic evolution process of STFN. The results showed that: (1) In the past 10 years, Shanghai tourist market had experienced a process of evolution from stable and orderly to short-term fluctuation and then gradual recovery, and the year of 2019 was the turning point of tourist flow network evolution. (2) The small-world and approximate scale-free characteristics of STFN were verified, and the network changed from disassortative to temporary assortative, showing a development trend of external expansion and internal separation. (3) While the centrality indicators of tourist flow network remained stable as a whole, the attention to cultural nodes was also increasing with the emergence of new nodes; (4) In terms of spatial connection, new popular nodes emerged and the relationship between them and the surrounding nodes was strengthened; (5) The spatial pattern of tourist flow network presented an inverted “V” shape and gradually expanded to southwest and southeast, forming a network with core nodes as the center and radiating outward. At the same time, newly emerging nodes at the periphery had formed relatively independent clusters.