This paper presents an integrated framework for hotspot mining and route optimization in scenic areas using tourist trajectories and behavior recognition. We propose BTOM, a probabilistic Behavioral Trajectory Optimization Model that jointly considers travel cost, hotspot related objectives, and behavior driven satisfaction with an adaptive feedback update to handle dynamic preferences and changing environments. We further develop BDPOS, a Behavior Driven Path Optimization Strategy that models the scenic area as a dynamically weighted graph and incorporates prediction of future crowd density and behavior distribution to proactively revise route recommendations. By coupling hotspot discovery with behavior aware, congestion sensitive routing, the framework improves both hotspot identification, and navigation quality. Experiments on multiple datasets show that our method consistently outperforms representative baselines in hotspot detection and path planning, while remaining scalable and adaptable across diverse scenarios. The proposed framework provides a practical solution for intelligent scenic area management and can support improved visitor experience and resource allocation and overall tourist satisfaction.
Qingbo Shi (Sat,) studied this question.