Analysis and application of reservation of scenic spot based on relationship graph

In recent years, the reservation data of scenic spots in Gansu Province has increased dramatically. This paper cleans and integrates these scattered data, refines the correlation relationship between them, constructs a relationship map, and inputs it into the constructed recommendation system, in order to achieve the accurate recommendation for tourists to reserve scenic spots. Firstly, this paper constructs the relationship map between tourists’ characteristics and scenic spots based on the real data of scenic spots reservation in Gansu Province. Next, the paper divides the data set into training set and test set to train the model of recall algorithm. Then, the paper compares the clustering effect of different clustering algorithms to find out the most efficient algorithm to get Top-N recommendation. Finally, this paper uses two sorting algorithms to sort the scenic spots in the recommendation list respectively, and thus achieve the scenic spot recommendation for tourists.