Urban Tourism Hotel Recommendation Model Based on Geographic Spatial Machine Learning Algorithm and Spatial Route Planning

Based on the analysis of the cost of urban traveling and the constrained tourism demands of tourists, we construct an urban tourism hotel recommendation model based on geographic spatial machine learning algorithm and spatial route planning. Firstly, based on the constraints of urban geographic space, an improved DIANA clustering algorithm is constructed to spatially cluster the scenic spots in the city and obtain a local distribution model of the scenic spots. Secondly, taking the scenic spot cluster as the research scope, a spatial route planning algorithm is constructed to output tour routes starting and ending at a certain hotel, and the optimal complete binary tree algorithm is used to output the cost priority of each route, thereby obtaining the hotel recommendation priority. The experiment proves that the proposed algorithm can effectively output the clusters of scenic spots under geographic spatial constraints, and recommend hotel accommodation with lowest traveling cost for tourists within the clustering range, providing the decision support for tourists’ tourism planning.