The development of mobile clients and social media provides a large amount of big data information, such as trajectories and photos, for tourist behavior research. This paper takes the trajectory and interest photos of the “two step road” platform as the data source, takes the Dianchi Lake and surrounding areas of Kunming as the research object, and analyzes the temporal and spatial distribution characteristics, and landscape preference of tourists through GIS kernel density analysis, spatio-temporal path analysis, and content analysis. The results show that: (1) Tourists around Dianchi Lake and its surrounding areas show the characteristics of “two peaks and one valley a year”; the density of tourist trajectories increases from “ring” to the northeast area year by year, and the interest points of photos also change from “single core” to “Multi-core” transformation, but most of them are concentrated in the northeast region, and the southwest region is weaker, with uneven temporal and spatial distribution. (2) Tourists have their own preferences for natural and cultural landscapes, and photo elements also show a certain tendency to combine, but Insufficient shaping of cultural landscape. (3) The research area has formed a “hot, medium and cold” spatio temporal distribution area, which has a certain connection with the landscape preference tendency of tourists’ environmental perception in the process of tourism. The study found that the overall tourism development around Dianchi Lake and its surrounding areas is unbalanced. It is recommended to highlight the image of tourism subjects, integrate regional tourism resources, adjust the structure of tourism products, strengthen the spatial complementarity of characteristic resources, and build an integrated spatial development pattern. The study aims to explore the hidden tourism information behind the big data to promote the holistic development of tourism around Dianchi Lake area from the perspective of tourists.