Abstract: At present, the evaluation of tourism resources had a strong subjective color. There were many variables and strong nonlinearity in each evaluation model. BP neural network model used MATLAB calculation program to evaluate tourism resources, which can avoid the influence of subjective factors on the evaluation results. By using the methods of literature comparison, theoretical analysis and expert consultation, the evaluation index system including 7 first-class indexes and 16 second-class indexes was constructed. On this basis, the evaluation model was designed, and the error value in the evaluation model was analyzed emphatically. The evaluation model was applied to the tourism resources evaluation of several major cities, and the evaluation results and error range meet the requirements.

Key words: BP neural network, model, error