| After decades of reform and development, Jiangxi Province as a result of rapid population growth caused a lot of problems, on the other hand, the population structure, distribution and other aspects of the rapid changes have taken place; these led to the population and resources, social, environmental, economic causes of conflict, especially in the province also belong to the less developed regions, due to the continuous growth of the population, which greatly affected the level of economic development of a series of problems in our province, and employment; short time to curb population growth problem is not possible, so to predict in advance of the population is imperative holds many lessons for the distribution of social resources, education, health, social welfare, employment and other issues ahead of time to prepare.For now, the model on population projections, there are gray system model, improved artificial neural network model, Leslie matrix model, logistic population growth model, time series model, exponential, linear regression, support vector machines model etc; these models can be used to predict the population, each model has a lot of scholars, but what specific model, there is no uniform standard; paper uses a logistic population growth model, time series model, gray model and Jiangxi population combined model to model predictions.This paper is to analyze the excessive population growth, social problems, followed by the use of logistic population growth model, ARIMA(p,d,q) model, GM(1,1) model to model and predict the population in Jiangxi, compare various models advantages and disadvantages, and finally for the lack of GM(1,1) model is proposed to combine modeling with GM(1,1) model and BP neural network model, and use genetic algorithm to optimize the weights and thresholds of BP neural network, the formation of gray GA-BP neural network model, the results indicate that a combination model can cover more information, more accurate prediction accuracy. |
PDF Full Text Request