Spatial Poverty And The Differentiation Mechanism Of Yuanzhou Based On GIS And BP Neural Network

Poverty is a global problem, what confused human and social development. And it is also a challenge for developing countries to face with during the process of sustainable development. The development of poor areas is a very important and urgent problem, in the process of our country is building a moderately prosperous society at present stage.Ningxia hui autonomous region is a minority autonomous region in west China, and poor areas are mainly distributed in the central arid zone and the southern mountains (i.e., distributed in Liupanshan concentrated special difficult areas), the region’s ecological is fragile, land is barren, perennial drought and less rain, soil and water loss seriously, species and natural disasters occur ed frequently, population, resources and environment are unbalanced, and at the same time, restricted by social, economic, history, etc, it has brought unprecedented pressure to the poor areas in a new stage.The article based on the spatial poverty and its related theory, regard Yuanzhou as the study object, to administrative village as the basic unit, building a space poverty factor index system that contains30indicators, from three aspects as nature, society and economy, selecting seceral dynamic data between2010and2013years. First, using Pearson correlation analysis to distinct the poverty leading factors and poverty elimination factors, then using OLS regression analysis to quantify the influence degree of various factors on the poor, and with the combination of GIS and BP neural network, calculating for each space poverty index, and showing the spatial distribution on the figure, with the use of GIS’s visualization features, revealing Yuanzhou’s space pattern of poverty and its differentiation mechanism, putting scientific and reliable theoretical foundation to Yuanzhou’s poverty alleviation and development, and finally get the following conclusion:(1) Through the Pearson correlation analysis and OLS regression estimate analysis, the results it is concluded that:By means of Pearson correlation analysis, we can see that the natural environment is one of the main factors affect the original state poverty, at the same time, the social environment adds fuel to the poor; Economic environment is alleviating and eliminating regional poverty, with OLS regression estimate analysis, quantificate the effect of every factors to the poor.(2) Use GIS spatial expression and based on BP neural network, the space simulation analysis result as follows:Villages with higher natural poverty index are mainly distributed in the north and northeast region, villages with higher social poverty index are mainly distributed in the southwest, economic poverty elimination index have unobvious distribution, but it is overall on the rise.Combined with the analysis and research of before, we can finally put forward poverty reduction strategies and suggestions in accordance with the original state of regional characteristicsof Yuanzhou:(1) In those with higher natural poverty index areas, we must to strengthen environmental protection and soil water loss management, at the same time to strengthen the management of ecological environment construction and pollutants, as well as the construction of basic farmland and drinking water safety. (2) In social environment regulation, implementation of the ecological migration engineering, for social poverty index higher region, we must to raise the level of basic education, and intensify the poor employment skills training; At the same time to speed up the development of medical and public information construction of the network traffic and. cultural information(3) Yuanzhou state as the key development zone in southern Ningxia, we need to implement the strategy of the regional central city, for those villages with lower economic poverty elimination index we can strengthen the training and the promotion of science and technology, and develop the characteristics industry and agriculture.