Research on User Behavior Information System of Tourist Attractions Based on Data Mining

In this paper, a new A-MML (Apriori-means Multilayer) model is proposed, which combines the innovative data mining method of multilayer perceptron (MLP), K-means clustering and Apriori algorithm to deeply study the user behavior information system of tourist attractions. Firstly, the model uses K-means clustering algorithm to classify user behavior data and identify different user groups and their characteristics. Then, multilayer perceptron (MLP) neural network is used to further explore the deep behavioral patterns of each user group, and automatically extract key features to improve the prediction accuracy. Finally, the Apriori algorithm is used to mine the association rules between user behaviors and find the potential relationship between user preferences and consumption patterns. In the experimental part, we compare A-MML with the traditional system model. The results show that A-MML model can not only achieve accurate prediction of user behavior, but also provide personalized service recommendation and optimization strategies for tourist attractions, so as to improve user experience and satisfaction. This fully verifies the feasibility and effectiveness of A-MML in the user behavior information system of tourist attractions in data mining.