Urban tourism competitiveness evaluation system and its application: Comparison and analysis of regression and classification methods

Under the background of economic transformation, tourism is one of the most dynamic and promising tertiary industries. How to improve the competitiveness of tourism has become a new idea for industrial upgrading in various regions. This paper builds an evaluation system consisting of five major aspects, based on the data of 75 tourist destinations; then, uses cluster analysis to classify cities, and uses logistic regression, SVM and random forest methods to predict the tourism competitiveness of sample cities and compare the advantages and disadvantages of the two methods - classification and regression. From the results of the empirical test, the results of the classification method are generally better than the results of the regression, and in the classification method, the results of the SVM are better than the results of the random forest. In this case, the SVM model gives full play to its ability to solve the problem of nonlinear classification.