Prediction Of Population Classification In Yunnan Province Which Is Based On BP Neural Network Combination Model

The population issue is always the key problem of the development of human society, the change of the population has a directly important influence for the future economy and the society development.Yunnan Province is a big province which has a large number of Minority and total population. To solve the problem of the Minority population is good for the economy development of Yunnan Province,the social stability and the national unity.The population policy is a direct way in regulating and controling every index of population.It plays an important role in economic and social development.Perfect forecast for the population has a direct relation with the scientific policy-making for the population issue,so to find a more scientific,more reasonable and more accurate population forecast model is am important research topic for the academic community and the government to make decisions.The total population of Yunnan province and the Minority population have changed a lot since the 21 century,and the number of the ethnic population scale is uneven.Under the new society environment,people have different points on how the population for Yunnan province and population for the Minority change. This article tries to give a medium and short term prediction for the total populoation of Yunnan province and the population of the Yunnan Minority with the GM(1,1)prediction model and the Logistic prediction model and also The BP neural network combination forecast model,researching the main feature of the population of Yunnan province to provide a a reference for the Yunnan province government.The GM (1,1) prediction model and the Logistic prediction model and also The BP neural network combination forecast model are the common population prediction method. This article fisrt introduces the theory part of the The GM (1,1) prediction model and the Logistic prediction model and also The BP neural network combination forecast model,and then combines with previous theory and practice,This article has built a BP neural network combination forecast model basing on the GM (1,1) prediction model and the Logistic prediction model which both are single prediction mode thinking about the advantage of the combination forecast model. We has selected the changes of population in Yunnan province in 1989-2014 and combined with the characteristics of each model and taken the appropriate parameters of population in yunnan province, Yunnan minority population and a handful of population of ethnic population in Yunnan province to set up the population forecast model which uses the numerically simulatiton,to analysis the relative error of the prediction results respectively. With the Contrasted analysis of each module Type according to the prediction accuracy standard, We choose the highest prediction accuracy model respectively for yunnan province total population, minority population, and handful of ethnic population in 2015-2030 population forecast in the future.The results showed that the BP neural network combination forecast model has the optimal prediction precision. By 2030, the population, the ethnic minority population and a handful of national population in yunnan province will respectively reach to 50.0503 million,18.329 million,390600.