Study on Prediction of Urbanization Level Based on GA-BP Neural Network -Taking Tianjin of China as the Case

In this paper, we establish evaluation index system of urbanization level, The three main factors and comprehensive factor score are obtained by factor analysis, we impot the BP neural network improved by the genetic algorithm. Urbanization level three variables prediction model and single variable prediction model are established respectively, Through the comparative analysis of different forecast models found that modified BP neural network three variables prediction model by genetic algorithm is superior to other prediction model in the aspect of nonlinear fitting capability and prediction precision, finally, we utilize the model to make a short-term prediction for the urbanization level of Tianjin.