City Air Quality Impact Factor Analysis And Empirical Study Based On Data Mining

In the middle of the last century, the industrialization advancement sped up, urban scale expansion and population sharp growth, so that the air pollution environmental damage event happened many times in industrially advanced country's big city, thousands of person pathogenesis or got killed, severely impair human body health and ecological equilibrium. The people start to realize that is restricting the human economy and society's healthy development enormously to the atmospheric environment quality. In recent years, our country air pollution prevention has made the very big progress, but as a result of each kind of reason, the situation our country atmospheric environment faces was still stern. For coordinated development of the atmospheric environment and social economy, research on the urban environment air quality impact factor and the influence mechanism, is very essential, also improves the actual problem which the urban environment quality urgent needs to solve.The question of the modern urban environment air quality management is how to manage the data resources effectively and mine the rich information which contains in these data, displays the information potential and the value fully, promotes the urban environment air quality management level. The data mining technology is effective plan to solves this question. The data mining method is the quite novel complicated system modelling method emerged internationally in the recent years, had very strong modelling ability. Introducing data mining to influencing factor analysis of the region environment quality evolution can study the primary factor of environmental variation by using the computer, and carry quantitative analysis. This article take the Wuhan environmental aspect bulletin from 1997 to 2005 and the statistical annual data as a foundation, first has carried on the quality synthetic evaluation on the Wuhan air quality, based on the analysis of city air pollution's influencing factor, innovatively introduced neural network and genetic algorithm to the air quality impact factor system research, has established genetic algorithm-neural network model of the air pollution influencing factor, used the MATLAB modelling software, has carried on the scientific quantitative analysis to the Wuhan air pollution influencing factor.The article is divided five parts. The author has outlined the air quality impact factor domestic and foreign literature in the introduction, has expounded this article research background, the paper frame, the research content and the research technique. The second chapter first introduced the Wuhan air quality function regionalization, has studied synthetic evaluation method of air quality, used the SPSS's Principal component analysis method to appraise the Wuhan air quality change, has analyzed the city air quality influencing factor systematically. The third chapter mainly elaborated data mining model of this article. This chapter first introduced the basic BP neural network algorithm, then unified the genetic algorithm and the neural network, have constructed the genetic algorithm-neural network model, finally has given the neural network variable selection algorithm and the Grey correlation. The fourth chapter has conducted the empirical study to Wuhan's air quality impact factor, and has carried on the contrastive analysis with the Grey correlation result. Finally, proposed the key measure of improving the Wuhan air quality.The present paper has mainly adopted the qualitative analysis and the quantitative analysis, has unified the theoretical model and the empirical analysis. The qualitative analysis is refers to the mass of atmosphere evaluating indicator and the influencing factor research and selects the concrete target, the quantitative analysis is refers to atmosphere quality synthetic evaluation and to the air pollution influencing factor excavation computation. The theoretical model is refers to the construction of air pollution influencing factor anatomic model based on the data mining, then carries on the empirical analysis to the typical region's target data. We not easily carry on deep analysis to the environmental information by using traditional the method, using the new research technique is challenging for the region environment economy complicated system modelling, has the practical significance to the environment economy's sustainable development. The present paper broke through two difficulties, first,based on atmosphere quality synthetic evaluation foundation, the auther has analyzed the air pollution influencing factor systematically; Second, constructed has inherited the neural network mass of atmosphere influencing factor anatomic model primarily, and has conducted the empirical study to the typical region. Its research results provide the policy-making basis for the formulation science's environmental protection and the pollution government measure and the environment economy's sustainable development.