Abstract
This paper chooses a number of indicators such as per capita GDP and economic level. First of all, the box diagram method is used to remove the outliers in the data, and then the Box-Cox transform is used to normalize the data. Then calculate the correlation coefficient between features, and finally determine 15 features for the construction of the model. On the basis of BP neural network, using GA algorithm to optimize the weight setting, a GA-BP model is established to predict the prices of different types of forecasting objects. Combined with the search ability of genetic algorithm optimization algorithm and the learning ability of BP neural network, the advantages of both are fully utilized to get better regression prediction results. The result of GA-BP is better than that of BP neural network, which shows that the prediction accuracy of the model is very high, in order to provide some reference for the prediction of other fields.