Evaluation Of The Level Of The Development Of New Urbanization In China Based On Particle Swarm Optimization And Support Vector Machine

In this paper, with the connotation and the development model of new urbanization as the theoretical basis, based on the Support Vector Machine(SVM) and Particle Swarm Optimization(PSO) as the guidance of methodology, combining qualitative analysis with quantitative analysis, normative research with empirical research, to evaluate the level of development of new urbanization in China. According to the results of empirical research, divides Hebei, Shanxi, Inner Mongolia and other 27 provinces into four regions and to horizontal compares the current situation of the development of new urbanization these provinces. Then access the suggestions and countermeasure about accelerating the construction of the new urb anization and promoting the development of new urbanization in China.The first chapter sets forth selection background and purpose and significance of the research, analyzes the significance of promoting the development of new urbanization for the overall development of China. Secondly, according to the time sequence, the chapter summarizes the connotation and the development model of new urbanization at home and abroad. Then introduces the research methods used in this thesis and the innovation, and finally to show the research ideas of this article in the form of technology roadmap.The second chapter describes and analyzes the current situation of the evaluation methods and the calculation methods about the development level of new urbanization in China. Through comment on the evaluation methods, summed up the relevant problems about the index system of the new urbanization in China, to lay a realistic foundation for the construction of index system in the following chapters. Then analyzes the shortcomings of the current calculation methods of the new urbanization, explains the advantages of particle swarm optimization and support vector machineThe third chapter strictly follows the basic principles of establishing index system and references research achievements about evaluation of the new urbanization form domestic scholars. And then standardized more than 1600 datum of data of Hebei, Shanxi, Inner Mongolia and other 27 provinces in 2012, and apply SPSS18.0 to the datum after standardization. After that, outputting significant and correlation matrix, and according to the matrix to select index form the primary index, and then constructs the evaluation index system of the new urbanization in china. The fourth chapter separaetly expounds the principle of support vector machine and particle swarm optimization and describes the operation process of PSO-SV M, for emphasizing the scientificity and rationality of the method. In addition, based on particle swarm optimization and support vector machine to evaluate th e level of development of new urbanization in China. In this chapter, according to the evaluation method, established index system and the datum after selectio n to establish the evaluation model on Matlab platform and use program langua ge operation. The program running results show, the mean square error is 0.00565827; and correlation coefficient curve and the real value is 99.7104%, which shows that the regression results are scientific, output the penalty factor the bes t C and kernel function are respectively 10.0179 and 0.0527. Finally, according to the simulation results of the level of development of new urbanization in 27 provinces, the chapter evaluates and analysis the results, and puts forward the corresponding countermeasures and suggestions, in order to provide some sugges tions for promoting the new urbanization construction in China.The fifth chapter summarizes the research of this thesis and the related conclusions are as follows:(1) the level of development of new urbanization in the eastern region is higher and the advantages are more obvious than others;(2) the level of development of new urbanization in the central region is lagging behind the other provinces;(3) the level of development of new urbanization in western area presents the ladder-like distribution. Secondly, looking ahead about the research direction of this topic from two aspects of evaluation index system and calculation method.