Research on Fusion Algorithm of Tourist User Location Trajectory Based on Multi-feature Space

The location key points of tourism users contain a large number of multidimensional information, such as user behavior characteristics, preference habits, which provides the tourism industry with a foundation for identifying the locations of interest to users. A single spatial position characteristic cannot distinguish the trajectory category with precision. This paper integrates the trajectory mean, trajectory length, trajectory direction, and velocity features based on a single spatial position feature to map the position trajectory into a multi-feature space. A multi-feature space-based travel user location trajectory fusion algorithm is proposed and compared to the conventional trajectory fusion algorithm using the user’s online travel text information and offline movement trajectory information. Experiments with simulation demonstrate the algorithm’s high computational efficiency, algorithm stability, and error convergence, as well as its ability to identify the user’s interesting location and make travel recommendations based on that location efficiently and comprehensively.