Analysis and Prediction of Regional Influencing Factors based on GRA-RBF Algorithm

Abstract

On the analysis and prediction of regional influencing factors. In this paper, under the three aspects of internal factors, international factors and investment, Beijing’s data in the past 10 years are used for study. In view of the ambiguity and incompleteness of the data information, the gray comprehensive correlation analysis (GRA) is used to identify the influencing factors, and the radial basis is further used. The radial basis function (RBF) neural network explores the potential relationship between different factors to achieve the purpose of predicting import and export trade. The results show that: (1) based on the contribution analysis algorithm, it can be found that the influencing factors affecting Beijing area, (2) 70% are used as learning samples for ANN training, 15% as validation set, and the remaining 15% as test set. Finally, it is predicted that the trade volume of Beijing will increase in the coming years.