Abstract Despite the importance of tourist mobility for urban tourism, mechanisms as to how these tourist activities shift throughout the day remain scarcely explored. This study fills this gap by providing a three‐step analysis framework. First, aggregated trajectory data were transformed into a directed weighted network. Then, a new community detection method was proposed to capture the spatial interactions, and two approaches were employed to quantify the evolution patterns of tourist mobility. Finally, the capability of the framework and its potential applications were explored and implemented through a case study in Xiamen, China.